Title of Invention

CONFIDENTIAL BLOCK TRADING SYSTEM AND METHOD

Abstract

ABSTRACT "CONFIDENTIAL BLOCK TRADING SYSTEM AND METHOD" A method for facilitating trading of securities over a computer system comprises electronically receiving a buy or sell order for a security from a first user; determining that the order is reasonably priced; transmitting to a second user an electronic notification that a reasonably priced order for the security is present, but without notifying the second user of the side of the first user's order; receiving an order from the second user, wherein the second user's order is a contra to the first user's order and sufficiently aggressive in price to cross the first user's order; and executing a trade comprising the first user's order and the second user's order at the limit price of the first user's order. Also described is an electronic system for facilitating securities trading that comprises a facilitator module, a financial information exchange server, a transactional database, and an analytics server.

Full Text

CONFIDENTIAL BLOCK TRADING SYSTEM AND METHOD Cross-reference to Related Applications This application is a continuation-in-part of U.S. patent application no. 09/870,845, filed May 31, 2001, which is a continuation-in-part of U.S. patent application no, 09/750,768, filed December 29,2000, which is a continuation-in-part of U.S. patent application no. 09/585,049, filed June 1, 2000. The entire contents of each of the above-listed applications are incorporated herein by reference. Field of the Invention The subject invention relates to a method for managing information to focus block-trading interests into a confidential crossing book, and executing block trades over a computer network such as the Internet. Background In public securities markets, market mechanics and trading psychology create barriers to efficient information dissemination and price discovery. A market participant's decision to reveal information regarding a large trading interest typically represents a tradeoff between confidentiality and liquidity. As used herein, the term "market participant" refers to any person or firm with the ability to trade securities; examples of market participants include broker-dealers, institutions, hedge funds, statistical arbitrage and other proprietary trading operations, and private investors trading on electronic communication networks (ECNs). By publicly revealing the details of a significant active buying interest, for example, a marrket participant assumes the risk of adverse price action — other traders may view his buying interest as an mdication that the stock is worth more than its current price. Other maricet participants that had legitunate selling interests, and market makers that seek short tenn trading profits, will "fade" their offers (become less aggressive sellers), resulting m an increase in the market price. There is also an empirically demonstrable risk of adverse price action due to "front running" (buying activity by other market participants in anticipation of price movement resulting from the large revealed order). Confidentiality can be maintained to some degree fay slicing the large order up into many small pieces to avoid arousing interest, but this process either fails to fill the order quickly when the trader is most careful to hide its existence, or if when it is worked at a faster rate, infonnation about the existence of the order leaks out leading to mucn the same effect as the above mentioned order display problem. h either case, an economically efficient transaction is missed because the trading costs associated with disseminating information are too high. Trading large blocks on the New York Stock Exchange floor is also hindered by rules designed to protect the interests of floor brokere representing short term trading Laterests, and the proprietary trading interests of the specialists themselves. Floor brokers can participate with an histitutional order, taking an equal portion of the liquidity coming in on the contra side, to take a short term speculative position with the safety net provided by the larger histitutional order, or "penny jump" on the Institutional order by intercepting the contra-side liquidity with a one-cent better pricey rather than allow it to fill the histitutional order. Specialists utilize various methods including penny jumping to extract profits from information about Institutional activity; invariably these activities lead to adverse price movements and lower fill rates for the histitutional orders. Another known difficulty with trading large blocks is one that is somethnes called the "buyer's remorse" problem. When a trader places a large blodc and gets it filled quickly, the trader's natural reaction is to believe that the easy fill indicates that it was a poor trading decision — either there was a much larger contra whose residue will subsequentiy drive the price adversely, or the block's price ivas too attractive, leading to its quick acceptance. This "buyer's remorse" effect leaves the client unsatisfied and unlikely to place similar orders through this same channel in the future, smce the quick fill would be perceived as a failure of this channel to provide protection from a poor trade. One known approach to matching trading mterests and racecuting block trades while limiting information dissemioation is employed by the POSIT® matching system. The POSIT® system allows trading interests to accumulate and initiates a matching sequence at set intervals. Maricet participants place confidential orders in the system and are unaware of the amount or aggressiveness of other orders on the same side or on the contra side until the matching is released. This approach does not provide for any mechanism to notify traders that may have nearly matching prices but overlapping order quantities that they could achieve a trade with only a small diange in their price. Also, by being a black box it fails to provide information that would mdicate to traders that there is liquidity in the system in particular securities, so that they would know to pull back orders they may have placed elsewhere in order to participate in the POSIT® match. This approach further fails to enable the broadcast of information based on the activity in the system that may entice traders to enter orders when the system's participants have a higher than usual interest in a given security, this being an indication that the fill rates could then be expected to be higher. It also fails to provide the ability for a participant to initiate transactions at a time of his or her choosing. By pricing all trades at the midpoint, it also fails to provide any mechanism wherein a patient trader with a moderately sized block could draw a better price from a much larger or more anxious block order that would tend to drive the price adversely. Finally, it does not offer any ability for market participants to notiate or otherwise actively participate in price formation through a process that could lead to the discovery of a fair price for a blodc trade. Another system that attempts to manage information to facilitate large blodc trades is die LiquidN Trading System. USCTS of this system contribute data regarding their trading interests directiy from their Order Management Systems. When a user indicates an interest that matdies contra interests aheady expressed within the system, all matched users receive an invitation to enter a negotiation session. If this invitation is accepted, the trader is experted to be firm and negotiate in good feith towards a trade; those that play with the system but fail to negotiate in good faith and close trades are expelled from the system. In the LiquidNet system, traders leani the existence of counter interests brfore anyone has entered negotiations; while this leakage is useful to entice traders to enter a negotiation process, it also leaks mfomiation about orders in the participants' Order Management Systems to parties that may or may not have a firm intention to trade. Tie system therefore relies on "feir play" by members of the ndwork, which lunits participation to a country club of like-minded Institutions. This fails to provide a system where the side of a trading interest can only be revealed to parties that have a fmn, electronically executable contra order within the systan at a reasonable price, and through this rule permits thepartidpation of parties that have different incmtives and trading horimns, including sell-side finns and hedge funds. TheLJquidNetinfonnationmanagementmodel, by leaking mfonnation regarding the side of a participant's order without a firm Uabili to trade, fails to achieve a proper balance between the need for confidentiality and the need to share some information in order to focus block interest m time, as can be achieved through means that indicate to traders that there is trading mterest within the system without revealing the side of said interest until a trader has been exposed to a firm liability to trade. It also fails to provide the protection from gammg that can be achieved by a block crossing system wherein the only means to discover the side of a second participant's order is to place an aggressively priced large block order, which would then have a significant probability of bang executed. By allowing orders of all sizes, the LiquidNet system also (fails Co provide a solution to the buyer's remorse problem, wherein a buyer that has a laie order completely filled with great ease later fears that the contra had a much larger sell order whose residue is likely to drive the price down. Haiborside Plus offers another system to facilitate the crossing of blocla based on the management of trading interest information. In the Haiborside system, traders are asked to send indicatioiK of interest into their database when they have an interest in bujing or selling at least 25,000 shares of stock at the current midpoint price. These indications of interest are stored electronically and expire after a certain number of minutes, which depwids on the clirat's trading style (for example, hedge funds or brokers tend to trade quickly so their indications will exjiire after only a few minutes). When there is an lOI on both sides in the security, a sales trader is notified through a popup window and will proceed to call the customers to dose a trade. The mediation of a human trader is intended to ensure that the system is not abused by traders that would enter an lOI when in fact they do not have an interest to trade, since the damage to the relationship wouid make it less likely that this trader would get called again in the fiiture. Tliis system does not anticipate a trading system wherein fair play is enforced automatically by requiring traders to enter firm, electronically executable orders, that could either execute on arrival or be exposed to an execution from another party resident within such a system before the owner of said order recMves any information back about other traders' interests. It also fails to provide any means for automating the notiation process that is carried out by the sales trada". Since a sales trader is involved, it fails to pnavide the anonymity that traders expect from a computer system, where traders are enabled to negotiate directly with each other without leaking information to a third human being who may or may not be fully trusted. Finally, the Haiborside system fails to anticipate mechanisms to entice other traders to express their trading interest in a security when there is evidence of trading interest on one side only. Fbally, it fails to anticipate the possibility of showing infonnation about one-sided trading interests without revealing the side, for the purpose of focusing interest in the security at this time without leaking price-impacting information to parties that Iiave not expressed a fmn liability to trade. A fourth example of a system that attempts to manage information to facilitate block trades is the Liquidity Tradcer facility, deployed by the Nasdaq Stock Market and subject of co-pending U.S. patent application no. 09/870,845, tiie entire contents of which are incorporated herein by reference. Liquidity Tracker enables a user to enter a firm order to trade and show information about this order to likely counter-parties. In the Liquidity Tracker system, the counter-parties are identified using real-time access to clearing information that Nasdaq participants submit through the Automated Confirmation and Transaction (ACT) system. By selectively targeting likely contra parties only, the intention is to avoid leaking infonnation to other traders that could front run on the Institutional order or otherwise misuse this information for short term trading profit. However, since the Liquidity Tracker system only selects contra parties, the recipients will know if the first participant was a buyer or seller, i.e. the side of the order entry participant is not kept confidential. Some recipients of this information may be induced to stop or even reverse their trading direction instead of accepting the order, in the hope of seeing a better price later. While the level of confidentiality provided by the selection of targets is appropriate for somewhat larger orders than those that could be displayed on a public book, there are instances where traders with larger institutional blocks will feel uncomfortable with this level of confidentiality. In this enviromnent, there is an acute need for a confidential crossing system that enables users to place an order that will not be shown to anyone, but wUl nevertheless cause an information event that can help bring traders together at the same time in the same security. In order to be productive rather than destructive, information events in a trade negotiation process should avoid two problems: 1. The preference revelation probiem. Once a trader has displayed an interest to buy (sell) a fungible item, other participants that see this interest will increase (decrease) the value that they were placing on this item. In equity trading, showing a large buy (sell) order leads others to revise and pahaps cancel their intent to take a contra position. On the other hand, by not expressing a preference one misses the opportunity to trade altogether. 2. The buyer's remorse problem. When purchasing a fungible item, when the deal is closed with greater ease than expected, the buyer tends to assume that it was likely too good a deal for the other party. In the context of equity trading, when a large block order gets completely filled the trader would assume that the contra party's order was even larger and its residue would subsequently affect the price adversely as it gets worked in the marketplace. This problem is particularly acute for a sjtem that caters to large Institutions. Summary The systems and methods of the present invention provide novel solutions to the above-mentioned problems of institutional block crossing. A preferred embodiment of the system limits the revelation of trading preferences by not showing the side on which one intends to trade until the final steps of a negotiation. Upon entry of a reasonably priced order, the system sets the symbol ui an "active" state. This indicates to other participants that an order has been entered in this security, but without revealing whether it was a buy or a sell order. In the preferred embodiment the system determines prices to be "reasonable" if they are at least as aggressive as the passive end of a Blodc Price Range (BPR). The BPR is calculated based on the current market prices, recent volatility in the symbol, and liquidity, and shown to the trader on a Graphic User fateriace (GUI). In the prefened embodiment tiie BPR is calculated by predicting the price range that one is likely to see in the next 60 seconds, and this range is re-calculated at 60-second intervals. In an alternate embodiment a "reasonable" price is deemed to be one that is within a given number of cents (or fractions thereof) from the National Best Bid or Offer, for buy or sell orders respectively, on the time of order entry. A second participant who sees the symbol active flag and has an interest in tradmg the same security can elea to enter an order. If it is on the contra side and is sufficiently aggressive in price to cross the first participant's order, the trade is ececuted at the first participant's limit The buyer's remorse problem is addressed by requiring all traders to enter orders only in multiples of a large block quantity. This quantity is chosen t> be somewhat larger than the typical size of orders usually sent to broker-dealers. For the most liquid securities, for example, the system can be configured to use a large block quantity of 100,000 shares. As a result, it will be the traders' expectation that most orders in the system are for one imit of this large block quantity only. As a result, when a trader receives a complete fill on a 100,000 share order this does not indicate in any way diat the centra's order m the systan should be expected to be larger-most likely it is also a 100,000 share order and the complete fill is only a reflection of the system's mandated large block size, hi an alternate embodiment, orders can only be entered for this large block quantity and there is never a residue after a fill. In yet another alternate embodiment, the traders fiirther commit to not trade on the market for 30 mmutes after a trade is done, and agree to enable random audits to verify that the voluntary restriction is respected, or when verification is deemed to be necessary. Having reduced the likelihood of buyer's remorse, it is then possible to allow tradere to know by means of a "contra present" flag when there is a nearly matching contra order within the system. Since the sizes of the two orders are Hkely the same, the infomiation about the presaice of matching orders on both sides does not point to any adverse price movement in the near future: if both traders canceled out of the match they would each be trading on the regular market and theu" respective trades would balance each other out with no net price impact. Therefore, the information that there are contras present is consistent with an assumption that the current maricet price is a fair price to close a deal on a late block. To avoid the possibility that some traders may enter an order on an aaive symbol only to see the contras present flag and then immediately cancel, the Systran described in the preferred embodiment lets resident orders seethe contras present flag immediately as soon as a contra order is entered, but delajB showing the flag to the second participant that entered the contra order and shows this flag only after 30 seconds if the contras present situation persists. This gives the resident order a 30-second time window to either execute the newly entered order by increasing the price aggression, or cancel his or her order before its existence is revealed to the second participant. The first mover advantage to the resident order pro\ides an incentive to enter orders that will set the symbol status to "active", thereby enhancing the liquidity in the system, which in turn will draw other traders that have an interest in taking said liquidity. In a prefeared embodimoit, traders may enter orders at any price but the active symbol and contra present flags are only issued based upon orders that are priced within a system-announced "Block Price Range." The Block Price Range is determined based upon the National Maricet Best Bid and Offer and the stock's recent volatility, with the intention to provide guidance on a range of prices within which one would be most likely to execute a large block trade. For example, the Block Price Range can simply be taken to be the range fi-om the National Best Bid minus 5 cents to the National Best Offer plus 5 cents. If all traders know the current BPR, they know that entering an order within this range will cause the active symbol flag to be displayed to participants. A second participant who sees the active symbol flag and would like to buy a large block of stock can then enter a buy order at the Block Offer (top of the BPR); if the first order that caused the active symbol flag to be posted is still open and was a sell order, the second participant would then have the certainty of a fill. Conversely, if such an aggressive buy order did not fill, the second participant would know that either the first order had been canceled or it was also a buy order. If the second participant instead entered a passive order - for example, an order to buy at the Block Bid ~ then that order would most likely not get filled on arrival even in the case where die first participant was a seller. Instead, the first participant would see a "contras present" message and may elect to lower his or her price to lock in the trade. If the first participant takes no such action, after 30 seconds the second participant would learn that the first participant's order that caused the active symbol flag was in fact a passive seller, and may elect to increase his or her price to meet at a reasonable price in-between. As long as both participants are seeing the contras present flag, they can modify their prices until a trade gets done; the more anxious trader will be more likely to raise the price aggression level more quickly, leading to a natural price discovery process even in the absence of any displayed prices other than the BPR guideline. In the preferred embodunent, the traders in a contras-presait situation do not receive information about the price movements of their counterparties; tiiey are simply expected to price their orders at what they consider a fair price and hope that the other party will do the same. In an alternate embodiment, the traders see that the contra patty has inaeased or reduced their price aggression and may use this information to deiermifle that it is worth sunilarly raising their own price aggression level. If each trader does this in turn by a minimum 1-cent (or other pre-set) interval the two will meet approximately halfway between the original buy and sell prices. A schematic view of a preferred computer system embodiment of the present mvention is given in FIG. 1. The system preferably comprises a trade feciUtation system 100 (also referred to herein as Cloud9). Said trade facilitation system 100 is connected to participants through a communication network 80 such as the Internet, and a financial information exchange network such as a FDC Network 90. System 100 receives market data from a vendor 60 through the vendor's network 70. Participants access the system through a graphic user mterface (GUI) 20 installed at eadi participant's workstation. Orders are routed to an Execution Engine 50. Matching orders are traded automatically when received by said Execution Engine. Cloud9 comprises a Net Server subsystem 110 responsible for connectivity to each client GUI; a Financial Information Exchange (FIX) Server 120 to provide connectivity to back-office systems and client order management systems; and an Order Manager subsystem 130 responsible for implementing the order handling logic of the system as described herein. Facilitator module 140 creates infonnation events intended to focus interest and steer users toward a trade. The trade facilitation system 100 keeps track of the status of its orders in a transactional database 150. The price aggression of orders is evaluated with respect to a Block Price Range, which is computed by an Analytics Server 160; the price of an order can be (a) more passive than the BPR, or (b) aggressive, meaning that the price is either within the BPR or more aggressive than the BPR. Tlie Execution Engine 50 receives orders through the Communications Network 80 and electronically stores them in a transactional, fault tolerant database (the Book 5 J); it reports executions back to the trade facilitation system 100 over this same network. In an alternate embodiment, the Execution Engine 50 can be hosted locally as an additional component within the Cloud9 system 100, Participants use a graphic user interface 20 to enter orders into the system. When said order is executed, the results are reported for processing ro tiieir sponsoring broker 95 and to their own firm's Order Management Systems (OMS) 10 using the FIX protocol. AU orders are preferably required to be "Large Block Orders" which means thdr size must be a multiple of a large block quantity. For example, if the Large Block Quantity were 100,000 the GUI only allows orders that are a multiple of 100,000 shares. The orders can be Limit or better, or pegged to a ma±et indicator such as the midpoint between the best bid and offer on the National Market. All validated block orders are immediately routed to the Execution Engine 50, where they are stored in the Book 51. In an alternate embodiment, ordos of any size are accepted. Both Cioud9 and the Execution Engine 50 preferably process orders on a first come first served basis. The GUI 20 enables a trader who has an Interest in trading a large block in a given security to click on the security's symbol as displayed in a box on top of the main screen, to bring up an order entry dialog and automatically populate the fields of the order entry dialog in accordance with his or her pre-configured preferences. The trader will edit these fields as needed and press a button marked "Submit' to place the order. If a newly entered order matches a prior book order (the new order is in the same security, on the opposite side, and with a price that equals or betters a standing book order), the Execution Engine 50 automatically executes a trade at the price of the standing order, and electronically reports the trade to the Automated Confirmation and Transaction (ACT) 40 service for the consolidated tape, and to the sponsoring broker's clearing finn 30. Notice of the execution is also sent to the graphic user interface 20 to notify the trader of the trade. If on the other hand there is no matching order in the book, the new order remains stored in the book as a standing order. In a preferred embodiment clients can elect to have their orders sponsored by a third party broker, in which case the trades are also reported to the sponsoring broker's back-office systems 95 via FIX. In order to increase the likelihood oftrades, the system's Order Manager 130 is connected to a facilitator module 140, which automatically takes action when a newly-entered order does not find a match but is stored in the book 51 as a standing order. In a preferred embodiment, only ordas that are reasonably priced are subject to facilitation; a reasonable price range for trading large blocks (the Block Price Range) is computed by the Analytics Server 160 and posted on the graphic user interfaces 20 oftraders with an update every 60 seconds. The purpose of the facilitator 140 is to generate information events that will Kicourage others to enter orders in the same security, without leaking the price or side of the order. In a preferred embodiment, there are two possible action flows when the facilitator 140 is called into action, depKiding on the state of the orders as known within the system's database 150. In the first case, whai a newly entered order does not find a match, the state of the systGO. is such that there are no reasonably priced contra orders within &e system. A message is delivered to all parties that subsoibe to this security, indicating that the symbol subject of the newly entered order is now "active" within the system (if it was already active, this step is omitted). The graphic user interface 20 will display active symbols in an oraie-colored box 215 above the order entry dialog 205, as illustrated in FIG. 2. The system preferably enables a second participant who sees an active symbol and has an interest in trading a laie block in this security to bring up an order entry dialog by clicking on the symbol. The system preferably populates all the fields of the order entry dialog 205 in accordance with the second participant's pre-configured preferences. Tlie second ■ participant is acting with the knowledge that the symbol is "active." Therefore, he will understand that a first order was previously entered that caused the symbol to become active, and this fut order was priced within the BPR. Taking advantage of this information, the second participant might elect to submit a limit buy (sell) order at the block offer (block bid), to essentially guarantee a trade if the first order was on the contra side and is still open. In a preferred embodiment, once a symbol has become active it will stay active for at least 60 seconds; the symbol activity state is reviewed in conjunction with BPR updates and set to inactive only if it has been active for more than 60 seconds and there are no more reasonably-priced orders currently open within the system at the time scheduled for a BPR update. In an alternate embodiment the symbol's activity status is updated in real time to reflect whether or not there is a reasonably priced order within the system at this time. In another alternate embodiment, in addition to the color the GUI displaj the number of reasonably priced orders that are currendy open within the system. In the second case, the state of the system is such that there is already a reasonably priced order on the contra side, but it's not quite aggressive enough to meet the limit price offered by the newly entered order. In this case, the facilitator module 140 will immediately send a message to all participants with reasonably priced standing contra orders to let them know that there is now a reasonably priced contra order present within the system. The graphic user interface 20 displays the "Contras Present" message by highlighting the symbol in a yellow-colored box 225 above the order entry dialog 205, as shown in FIG. 2. The first participant, who caused the symbol to be active, will therefore see that there is now a contra present. The system preferably enables said first participant to re-price the order at an aggressive price and replace the order so as to Kcecute the second participant's order. In a preferred embodiment the system automatically proposes the aggressive end of the BPR as the defeult price to encourage a straty that will result in a trade; the trader can then modiiy the order parameters as desired. In an alternate embodiment the system enables users to configure an aggressive order type to be limited to the better of the BPR or a configurable number of cents (or fractions thereof) Bom the current National Market Midpoint price, and this aggressive order type is then the one that is proposed as the default in a contras present situation. In a preferred embodiment the user is allowed to enter an order that will cause the system to attempt to execute tlie contra's order by means of cliddng on the yellow box with the symbol and then pressing a single button labeled for this purpose. In an alternate embodiment the user mterface 20 lets the trader select a new price and press a button marked "Replace." Ifnooe of the traders with staadingorders raise their price aggression, and there are still standing contra orders after 30 seconds, the second participant that originated tfie contra order will preferably also learn that diere are standing contra orders within the system. TTie sjtem preferably displays this information by means of a yellow box, as described previously for the first participant; and said second participant may elect to increase his or her price aggression to close a trade. The 30-second delay is intended to discourage traders from probing the system to discover whether contras are present, and then attempting to cancel immediately thereafter. This safety mechanism is effective to the extent that the BPR is not too fiir removed from the current maricet, so that during the 30-second interval the newly entered order faces a real chance of being executed by one of the traders with a standing order. In an alternate embodiment there is no 30-second delay and all traders are simultaneously informed of the contras present situation, but there is a 5-second minimum time in force during which orders may not be canceled. In another alternate embodiment all users receive the contras present message immediately and there is no minimum time in force condition. In an alternate embodiment traders may configure the GUI 20 to automatically replace dieir order with an aggressive price when a contras present situation arises. In tiiis last embodiment if there are multiple orders within the system with automatic re-pricing instructions and cornpeting to execute the same contra order, the system executes orders in price-time priority: standing orders compete on price first to fill the incoming contra order, then if two ordera have the same aggressive price limit, on time of entry, with priority gomg to the first order entered. A preferred algorithm to calculate the Block Price Range (BPR) seeks to determine reasonable prices for trading a large block order within the next 60 seconds, depending on the volatility of the stock. In the preferred embodiment; this range is based upon the prices at which trades have been reported to the consolidated tape since the last time the BPR was updated, taking a symmetric price range up and down from the current midpoint of the National best bid and offer. This range extends up and down from the midpoint by an amount that is directiy related to the observed price volatility smce the most recent BPR update. FIG. 1 is a sdiematic view of a preferred computer syst«n embodirawit of the present invention, FIG. 2 depicts a prefen-ed GUI whein an order entry dialog is open. FIGS. 3{a) & 3(b) depicts operation of a preferred system and method, including interaction of the sjratem with electronic systems of participants and related firms. FIG. 4 shows the life cycle of a trade firom the clearing perapective, in a preferred embodiment FIG. 5 depicts steps preferably executed while processing a New Tidcet. FIG. 6 depicts preferred steps for processes a Cancel Ticket message. FIG. 7 depicts preferred steps for processing valid orders. FIG. 8 depicts a preferred watch list configuration dialog. FIG. 9 depicts steps of s preferred method for calculating BPR. FIG. 10 depicts a preferred GUI wherein watched securities are identified. Detailed Description of Preferred Embodiments The following is a functional description of preferred and alternate embodiments of the subject invention. Key features. Below is a preferred list of features of the subject invention: 1. Confidential crossing of blocks. Orders are routed to a matching engine that will cross orders with price time priority and report locked-in trades for clearing. The matching engine can be remotely hosted. 2. Large Block Quantity. All orders must be entered in a multiple of a large block quantity configured per security. The large block size deters gaming and mitigates the "buyer's remorse" problem, or fear of being on the wrong side following a complete fill, by encouraging users to use the same order quantity. 3. Active Symbol Flag. The system will push active symbol alerts when it determines that there is block interest on both sides of a security. This will focus order entry m a narrow time window and therdiy increase the fill probability. 4. Contras Present Flag. Traders with a reasonably priced order in the system will be alerted when a nearly matching contra order is entered. 30 seconds later, the participant with the new order is also alerted of the presence of a near match. The contras present flag focuses orders in price to maximize the probability of a trade. 5. FtX Connectivity to Badc-ofFice. Cloud9 supports mechanisms to deliver FIX executions to a Bu>ide Orda- Management System either directiy or through a FIX Service bureau. Supports end of day anonymity for the buy-side client with respect to the sponsoring broker. 6. FIX Ticketing application. Clients can place shares from their Order Management S>tem to Cloud9 (symbol, quantity and side), then work up to (he placed quantity using the GUI 20. The OMS can be updated to keep track of individual working orders, or only upon a fill, depending on the client's configuration. 7. API/GUI Access. Cloud9 may provide a house-branded trading front-end designed with input fi-om institutional traders, that prominently displays the active symbols for a watch list of securities of interest, and prompts the trader to close a trade when the contras present flag is set. The API may also be published to enable approved third parties to build access from their own fi'ont ends. The invention comprises a preferred architecture for integrating the system with the electronic systems of participants and related firms such as the sponsoring brokers or their clearing firms. An overview of this integration is represented in FIGS. 3(a) & 3(b). These figures illustrate the flow leading to a trade between participants C and E, described below. With standing orders already in the system, a new participant enters a contra order. The sjtem sends out the contra present flag; participant modifies the standing order to a higher price aggression level, resulting in an executed trade. The system preferably facilitates the trade in the above example through the following steps: 1. In step I a user enters an order into the system, preferably using a Graphic User Interface 20. 2. The system preferably assigns abrokerofrecordinstep 2 and validates the order against any credit limits based on the broker of record and customer ID. The order information is stored in a database 150. 3. In step 3 the Order Manager 130 routes tiie order to the execution engine 50, and ejqjects an order update message back from the execution engine mdicating whether the order was filled on entry, stored as a valid limit order, or canceled if the order was invalid. 4. Upon finding that the order did not execute or cancel m the execution engine, the OM 130 notifies the Facilitator Module 140 of the order mtry event. 5. The Facilitator Module 140 finds that there was a contra prraent in the system; it sends the residait contra the "contra present" message immediately and schedules an event to release this same message to the trader that just entered the new order after a 30-second delay. 6. lii step 6 the contra present message is delivered to the resident order. This is preferably done immediately. In an alternate embodiment the contras present flag is sent after a 10-second delay, to give the second participant time to try a more aggressive price before leaking information about the order. 7. Step 7 preferably occurs 30 seconds after the first contras present message was sent: the contra present message will be delivered to the new order, if the contra present situation persists. 8. In step 8, the resident order, a sponsored customer using a service bureau for their FIX connectivity to the system, receives the contra present flag on their GUI 200 and modifies the order to increase price aggression. 9. in step 9, the OM 130 routes the modify order request and updates the sUite of the order in the OM database 150. 10. ftistep lOtheOM 130 passes the modify onier message to the execution engine 50. 11. In step 11 the execution engine 50 determines thut (he new price aggression is sufficient to lock in a trade and preferably reports this trade to ACT with the clearing counter-party information. In an altemate embodiment the trade is reported for tape only and the clearing information is withheld until end of day to delay the tune when sponsoring brokers are notified of the trade. 12. The execution engine 50 preferably reports the execution to the Order Manager 130; OM 130 updates the state ofthe order in its database 150 and updates the credit consumed. 13. In step 13 the Order Manager 130 reports the trade to Net Server 110 and to FIX server 120. 14. Net SHver 110 and FIX Server 120 forward the execution message to their respective clients. 15. The second client receives the FIX executions through the service bureau. 16. In the last step ofthe trading process a drop copy ofthe exeaition is sent to the sponsoring broker(s). In the preferred embodiment this is done at the aid of day; in an alternate embodiment it is done immediately and the sponsoring brokers are directly inforaied of all trading activity they sponsor. Facilities The system preferably comprises tiie following facilities or subsystems: i. NetServer 110. TTie NetServer subsystem enables clients to access the system through a Graphic User Interface 200. This is preferably the only interface that can be used to aiter orders into the system. All users must either download the trading application (GUT), or develop a GUI that provides satisfactory visibility to the active symbol and contra present flags. Net Server 110 preferably supports connectivity to the trader front end via a published API. 2. FIX Server 120. The FIX Server is a computer system that allows clients to connect for the purpose of exchanging financial infonnation messages according to the FIX protocol; source code for a FIX server is available in open source code librari. The Cloud9 FIX Server preferably is configured to receive Tickets (New, Cancel, Cancel/Replace) and order status request messages. It will push out Executions including order updates and fills. It preferably supports direct connections with client Order Management Systems, as well as connections through a service bureau. Tlie FIX Server may also be used to deliver execution messages (with "done" status) to a sponsoring broker to set up after trade processing. Normally these messages will be issued at the end of the day, but clients preferably can also request the system to notify their broker of executions intra-day if needed. 3. Order Manager 130. The Cloud9 Order Manager processes orders and subsequent trading messages, passing them to the execution engine for storage and execution. It preferably receives BPR update messages from the Analytics Server 160 and maintains the price aggression flag on its orders, by keeping track of when the order is priced at least as good as the BPR, or Passively priced. 4. Facilitator Module 140. The order manager 130 preferably comprises a component in charge of fecilitating the entry of matching orders into the system by carefully releasing information about orders, as described herein. 5. Order Management Database 150. The order manager 130 preferably keeps track of the state of the orders in a transactional and fully recoverable Structured Query Language (SQL) database. It also contains reporting modules responsible for reportmg the trades to clearing as well as for generating daily mandatory reports to the corresponding Self Rulatory Organization (SRO). 6. Analytics Server 160. Tlie analytics server 160 receives a data feed that informs said server of the market activity state per symbol (start, stop, trading halts), the inside bast bid and offer for all Nasdaq and NYSE listed stocks (price, aggregate size and timestamp) and the last sale trade data from the National Market System consolidated tape. The server will preferably store this infonmation into a data cache and cany out analytical calculations to determine the Block Price Range. It is responsible for publishing to subscribing parties, per symbol, inside quote price changes, BPR update messages, and trading halts. 7. Help Desk. The system preferably comprises a Help Desk subsystem, which comprises a user interface such as a web GUI. Help desk operators can access tiiis intaface to add/remove/modify client fums, sponsoring brokers, or GUI accounts that are associated with individual traders. The help desk enables trading operations pereonnel to query the system based on an order ID or by firm and symbol, to trace the sequence of events tiiat follow a client's order entry into the system and give traders that wish to call the help desk a detailed response on the history of their orders through the system 8. Systems Operations Management The system preferably also comprises a systems operations console subsystem enabling a system operator to manage the fiinctional implementation of all Cloud9 facilities. Besides the common system operations duties, the system preferably also enables operators to carry out other duties including: security; auditing; managing batch processes for end-of-day and begin-of-day events; capacity planning; and managmg the Kititittnents for internal accounts (operator and Help Desk). 9. Matdung Engine 50. The Matdiing Engine 50 can be an Electronic Communications Network, or a third party elearonic trading book sudi as Nasdaq's SuperMontage facility. In an alternate embodiment the system itself comprises a matching engine as a separate component within the CJoud9 system. Preferred implementations of the above features are describad more extensively below. FIX Server. Cloud9 preferably uses the FIX protocol for back-office integration with the client Order Management Systems 10 and the sponsoring brokers when needed. FIG. 4 shows the life cycle of a trade from the clearing perspective. The clients are preferably able to choose between two models for using the system. The first model uses tickets to facilitate managing habJHty across multiple market destinations. Tickets are placements from the client's order management system that helps traders keep a tab on how many shares they are doing in each market destination (a market destination is a place where one can execute orders, such as the present system). The purpose of the ticket is to ensure that the aggregate number of shares placed across all destinations will never exceed the total number of shares of the client's overall order, as known to the client's Order Management System 10. When a trader uses tickets, the tickets that have been entered from the OMS appear on the GUI 20 and orders can only be entered against these tickets. The client that elects to use tick preferably will only be able to enter an order into the system if said order is in the same security, on the same side, and of a size not greater than a previously-entered ticket. The system preferably updates badc-office and third-paity systems by performing the folIowLag steps when traders elect to use tickets. 1. In step 410 a ticket is entered into the system 100 as a FIX Order with no price. Tie ticket carries a terminal ID that ±e system maps to a particular GUI. The trader sees die ticket show up on his/her GUI 2. In step 420 a trader enters an order against this ticket, which we will refer to herein as a "child" order. This child order is in the same security and side as the ticket, and for a quantity that is diosea to be the lesser of a trader-configured default order quantity or the tidcet quantity. The systan preferably validates the order and passes it to the execution engine. In the preferred embodiment the order entry events are reported to the OMS via FIX Execution mrasages with Order Status=New. In an alternate embodiment the client OMS will only receive Execution messages when the trade is completed (see below). 3. When a trade occurs against this order, the execution engine preferably reports to ACT for tape and clearing (step 430); then notifies CIoud9 of the trade. In the preferred embodiment the Cloud9 system then notifies both the GUI via API and the clients' Orda- Management Systems via FIX. 4. The system preferably enables the trader to initiate the allocation process before the end of the trading day (step 440 is executed at user's option). To obtain intra-day anonymity, the sponsoring broker is preferably not notified of the trade until the end of day, but the allocation process cannot be completed without notifying the sponsoring broker of the trade. To support end-of-day allocations, the GUI preferably enables the trader to first click a button to request that the systan report the trade to the sponsoring broker. This will prime the sponsoring broker's systems for the allocation process. In an altanate embodiment the sponsoring broker is informed immediately of each trade via a FIX Execution message. 5. In step 450 the system preferably s&ids a FIX Execution message to the executing broker, including the Cliait-ID. 6. The preferred system carries out the following (step 460) at the end of day. a. A FIX Execution message is sent to the executing brokM', including the Cliend-ID. This may dupUcate a similar message if it was already sent in step 450. b. A trade summary file is sent to the sponsoring broker's clearing firm. This contains the information pertinent to clearii, including the symbol, shares bought or sold, and contra party's MPTD. c. OATS reports give a summaryoftheordertrail and executions on behalf of the sponsoring broker. The system automatically sends the OATS reports to the sponsoring broker(s) for forwarding to the NASD. In an alternate embodiment it sends the report directly to the NASD. These file transfers are electronic and utilize the File Transfer Protocol (FTP). In the prefrared embodiment meis are enabled to select an alternate configuration that does not require the use of tickets, but instead allows traders to start the process at the ordffl- entry step 420. The remaining flow is the same as described above. In an alternate embodiment, the system 100 is hosted by the sponsoring broker for ali clients that use the system and comprises after-trade-processing subsystems, as can be easily implemented by those skilled in the art or are commercially available through specialized vendors. In this alternate embodiment there is no need for end of day anonymity and the broker execution reports can be dispatched immediately. The FIX Server 120 preferably supports a variety of networking solutions to provide connectivity to client Order Management Systems 10. For eicample the client's OMS can connect directly to said FIX Server; or it can communicate with FIX Server 120 through a FIX Service Bureau that fens out connections between a single FIX Server 120 and a plurality of clients. The FDC Server 120 will preferably reject tickets when they are invalid or when the client is configured not to use tickets; the ticket reject message preferably carries an mtelligible reject reason in the text field. If the rejection is sent because the order quantity was less than the Large Block Quantity for this security, the correct I3Q is preferably given for mfomiation in the message text in the reject message. In the preferred embodiment the FIX Server 120 preferably supports the following mcoming messages: FIXNew Order, Cancel Order, Cancel/Replace, and Order Status Requests. Outbound messages preferably include Order Updates relatir to child orders of a given ticket (a FIX Execution message carrying the Order-ID of the underlying ticket), if the client uses tidtets. Clients that do not use tickets will preferably receive fills only (FIX Execution messages). In an altranate embodunent cUents that do not utilize tidcets receive Order Update messages for each event relevant to an order entered via the GUI, including New Order acknowledge messages, so that the OMS cankeep track event-by-event ofthe total size being worked, as well as the size that has been fiUed. The FIX Server preferably supports a web configuration screen which enables an operator to set up and customize new client coimections without requiring a new software release. As mentioned above FIX buy-side clients can preferably be configured for either (a) FIX Executions only, or 0>) to support ticketmg and execution reporting. FIX Sponsoring Broker clients wUl preferably be configured to receive executions only. In an altemate embodiment it is possible to configure a sponsoring broker's FIX connection to receive a drop copy of every message sent to the sponsored client's OMS. In addition to the execution message sent on each fill, the configuration options preferably mclude an option to send a fmal FIX Execution message when an order is complete, with Order Status field set to "done", to indicate that the work on this order is complete. Order Manager. Ite Order Manager 130 operates as a message processor, hi the following paragraphs the processmg of various messages is described. The system preferably enables users to enter New Ticket messages. A New Tidcet message is a FIX New Order message witii Order Type = Market and no iunit price. FIX Ordera with a limit price will be rej ected. In an alternate embodiment this message may be submitted over an Application Programming Interface (API). llie following steps are preferably executed while processing a New Ticket (see FIG. 5), aftertheticket is received at step 510: • Validation. At step 520, the system preferably processes new orders by first validating the fields in the FIX New Order message. A ticket must have a valid client-ID, a trader-ID (that maps to a unique GUI), security and side (Buy or Sell) and be at least as large as the blodc quantity for this security. Cloud9 preferably accepts FIX Orders (tickets) identified as "Maricet" orders and "Not Held", and rejects tickets that have a limit price. In alternate embodunraits the restrictions on price and Not Held status are not enforced. A client preferably is only permitted to have one ticket per security on eadi side; a second buy ticket or a second sell ticket in the same security will be rejected if the prior ticket still had leaves. If the prior ticket was filled or canceled, then you can enter a second ticket; it will replace the first on the GUI front end. An alternate embodiment employs a fttint end that enables the display of multiple tickets per sscuiity and the client is permitted to place any number of tickets. Other business FIX order attributes will be disregarded. Rejected tickets are preferably displayed on the GUI with a reason code. • Processing. The system preferably stores valid tidcets in its database 150. The ticket is then displayed on the client GUI via NetServer ] 10 at step 540, and acknowledged back to the client CMS via FIX 120 at step 530. The table below gives required fields in the Ticket message in the preferred embodiment The systan preferably enables users to enter Cancel/Replace Ticket messages. A Cancel/Replace Ticket is preferably submitted as a FIX Cancel/Replace Order with the ClientTicketlD of a prior open Ticket and the ClientTicketlD of the new ticket. In an alternate embodiment this message may be submitted over an API. In the preferred embodiment, the system carries out the following steps upon receiving a Cancel/Replace Ticket message. " Validation. The system subject of the present invention preferably rejects the Cancel/Replace message if the new quantity minus already-filled quantity (new "Leaves") is less than the system Large Block Quantity for the symbol, or if it is less than the total amount of shares placed on the execution engine as open orders (leaves less than working). • Processing. If it accepts the request, CIoud9 preferably modifies the ticket size to be applied to any subsequent order entry events and pushes the change to the client GUI via NetServer 110. It preferably also acknowledges the change back to the client's OMS via FIX. The system preferably enables users to enter Cancel Ticket requests. A Cancel Ticket message is a FIX Cancel Order message with the ClientTicfcetE) of a prior open Ticket. In an alternate embodiment this message may be submitted over an APL The system subject of the present invention processes the Cancel Ticket message as follows (see FIG. 6)- • Validation. Tlie system preferably validates the fields of the Cancel Request message in step 610 by ensuring that it points to an open Ticket. • Processing. If there are open orders, in step 620 the system will first try to cancel all open orders associated with the ticket, then report any executions 630 that were done prior to the cancel with messages in flight (this "cleanup" process ends when all related ordera are either filled or canceled and the CIoud9 Order Manager's database agrees with the Execution Engine's database on the status of any fills). When any open orders have been deaned up, in stqj 6.40 the system wij] admowledge the Cancel Ticket with the results indicating the size that was canceled, if any. A successful Cancel Ticket request will be notified to the Client GUI via NetServer in step 650. TTie system preferably enables users to alter New Order messages. A New OnJer message is preferably submitted through the GUI or API and comprises a request to buy or lell a given number of large block lots of a given security at a limit price or better. In the referred embodiment the user may also select a pegged price limit, in which case the order mil be limited at the less aggressive price between the absolute limit price and the peed limit price. New Order messages are preferably handled as follows. Validation. The system preferably validates that orders are for a multiple of the large block quantity in a supported security. In an alternate embodiment the order can be for any quantity. Another step in the preferred order validation process is to verify that the order contains a price field. In an alternate embodiment this requirement is waived if tiie user has selected a pegged price, in which case a pegged order may float without limit. Tlie Peg types supported in the preferred embodiment are pegging to the NBBO Midpoint, or no paging at all. Alternate anbodiments support additional pegging attributes, as are known to those sldlled in the art. Tlie system preferably also validates that the side of the order is Buy or Sell. An alternate embodiment also supports short sell orders. The parameter PegOffset determines the number of cents (or fractions thereof) fhim the pegged value (NBBO Midpoint) where the order should be priced. TTie PegOfiset attribute is preferably disranded if the order is not pegged. The system preferably reports rejected orders to the GUI Client wiiJi a comprehensible error code. If the client uses ticketing, the validation step also preferably checks that the order does not exceed the size of the associated ticket and Js on the same side as the ticket. If the client entering the order has a credit limit with a sponsoring broker, the system preferably verifies as part of the validation process that the order does not violate said credit limit; the credit verification process will be described in greater detail below. In the preferred embodimeat IOC ordera are rejected as invalid, since tbey can be used to probe for information about a standing order that has caused the symbol active flag without showmg the contra present flag to the owner of said standing order. In an alternate embodiment IOC orders are accepted but the owner of a standing order can see a flashing Contra present flag (e.g., the symbol changes from orange to yellow briefly then reverts back to orange). In yet another alternate embodiment the IOC orders are accepted and do not cause any flags to be sent out, be it an active symbol flag or a coatras present flag. Order attributes. Tie system of the preferred embodiment expects to find the required fields in the New Order message as given in the following table. Processiog. TTie system preferably processes valid orders througfi the following steps (see FIG. 7). • After validation (step 710), valid orders are passed to the execution engine immediately in step 720. Tie response will indicate whether the order was Rected, or accepted and placed in the book as an open order, or has an execution pending. The status from the exeojtion eogme is reported to the GUI client via NetServer. • If the client uses FIX Tickets, the system preferably reports the order status bade via FIX as Accepted or Rejected (step 730); the execution pending status will be rqported as Open status in FIX (a subsequent lod:ed-in execution message will follow). The FIX order status update message carries the ID of the associated ticket. • If the Execution Engine indicates that the order is Open, the system will initiate a proofs 740 through the fiicilitator with the aim of increasing the likeUhood of a trade. The system preferably determines whethw the price is within the latest published value of the Block Price Range (PR), and if so, tags the order as "aggressive." For tiiis evaluation, pegged orders are priced using a Market Data Cache. tf the order is marked as "aggressive," the system preferably determines whether there are aggressive contra orders m the system. If there are, the Contras Present, flag is sent immediately to resident contras that do not already have it, and an event is sdieduled to re¬evaluate the Contras Present status after a delay of ContrasPresentDelay seconds. If both the first participant's order and at least one contra order are still aggressive and within the BPR at the occasion of this sdieduled event, the system sends the Contras Present flag to the first participant. ° If the order is marked as "aggressive" and the symbol is not already in an active state, set the symbol in an active state and generate an Active Symbol message. ' In step 750 the system waits for contra orders to arrive; each contra order is processed as in this flow starting at the validation step 710. Contra orders with an equal or better price preferably execute at the resident order's limit price. The system preferably enables users to enter Cancel Order messages. Cancel Order is a GUI/API message with an ClientOrderlD corresponding to a prior New/Replaced Order, In an alternate embodiment cancel order requests may also be received via FIX. Validation. If the order ID is invalid or points to an order that is not known withm Cloud9 as being Open, the cancel order request is rejected. Processmg. The system 100 preferably processes Cancel Order requests through the following steps. • CIoud9 OM 130 passes valid cancel order requests to the execution engine 50. • The Execution Engine 50 will respond with the shares that were canceled. The cancel response may be preceded by an eKecution(s). For filled or expired ordera the cancel request is rgeet. • Upon receipt of Ae successfo! cancel response om the execution engine 50, The order status update is pushed to the GUI Client 20 and the Analytics Server 160. If the client uses FIX Tickets, also report the status update via FIX. Release credit that was allocated to this order. Check whether there were contras with a "contra present" flag and remove the flag if the contra present condition is no longer met Note: a cancel order does not remove the active symbol flag. The system also enables the user to enter cancel/replace order requests to modify a prior order. In the preferred embodiment the Cancel/Replace Order is a GUI/API message pointing (CUentOrderlD) to a valid prior New/Replaced Order. In an alternate embodiment this message can also be entered via FIX. Validation. The CUentOrderlD must point to a valid order. The system preferably rejects a cancel/replace request if it points to an order that is already known within Cloud9 to be canceled, expired or filled. In the preferred embodimEsit the system supports pre-trade credit checking. Therefore, if the size of the order is to be increased through the cancel/replace request, the s>Btem first verifies that the new size would not exceed the credit limit on the user's account Replace order requests that fail credit validation are rejected as invalid with a reason code explaining the credit reason. Cancel/Replace Order processing. CloudP OM 130 processes cancel/replace order messages by passing them through to the execution engine, The execution engine 50 will respond with the order status including filled shares, rMnaining open shares, and average price. " Trade Message. The status update will be pushed to the GUI Client via NetServerllO. • If the client uses FIX Tickets, also report the status update via FIX. • If the cancel/replace modifies the size of the order, cancel and replace credit amount allocated to this order to reflect the new order. Trade Message. The execution engine 50 reports trades to the Order Managw 130 via a Trade Message. Each Trade Message fi-om the Execution Bigine rorts a lodted-in trade involving exactly two orders. The Execution Engine 50 will match individual orders against a book of contra orders (one-to-many match). There can be multiple individual trades following a match check, each of which will be processed mdependently. Since a trade is a single traasactiojo, bothlegsofthetradeshouldberortedasa tmit. The preferred embodiment of the system expects to find the following fields in a Trade Message. The system preferably enables traders to cancel all orders they have previously placed into the s>Btem with a single request, in the event of an emergency. CancelAllCIientOrders is a message from Net Server requesting cancellation of all orders that have been entered by a given GUI. In an alternate embodiment this message can also be delivered via FIX. Since the execution engine 50 does not know the customer ID, Cloud9 OM will process the cancel-all message by individually canceling every order associated with this client that has an Open Status, processing these as individual Cancel Order requests. In order to keep track of the price aggression of orders that are associated with contras presait or active symbol indicators, the Order Manager 130 in the preferred embodiment subscribes to BPR Update Messages from the Analytics Server 160 in securities for which it has an opea order. Whenever the BPR is refreshed, OM 130 reviews the "aggressive" flag on open orders and updates the flags as follows. if there is a change in the price aggression flag on an ordo", re-evaluate if there is a contras present situation for all orders in this security and update contra present flags to ON state when there are coatras present for an order that are priced at least as aggressively as die BPR, or OFF if there are no longer eligible contras present. If this results in a diange in the Contras Present Status on an order, the system preferably sends a Contras Present Message to affected GUI cliaits. ITie system of the preferred embodimait also updates the active status of an order based on BPR update messages. If the symbol is in an active state, has been in this state for at least 60 seconds, and tha"e are no longer any orders within the BPR, the symbol is set to an inactive state and NetServer sends an Active Symbol message with ActiveSymbolSlatus = Not active. Unsubsoibe to BPR update messages. In an alternate embodiment the active symbol status is updated oo any BPR update message and may be set to an inactive state even if the symbol has been active for less than 60 seconds. In yet another alternate embodimozt the active symbol is updated on every change to the National Mark« Best Bid and Offer prices and every order entry or caacdlation event to ensure that the active symbol flag always indicates that there is a live executable order in the system at this time, priced at least as aggressively as the passive end of the BPR. ExecutioD Enpine. lie execution engine 50 in the preferred embodiment is a decoupled system that communicates with the Order Manager 130 through a communicatioQ network 80 such as the Internet. In an alternate embodiment the Execution engine 50 is hosted in the same facility as the Cloud9 system 100 and the two systems communicate over an intranet. In antrfher alternate embodiment the Execution Engine 50 is a component within tile Cloud9 system 100; in all these embodiments the fimctioaality of the Execution Engine 50 is as described below. The Execution Engine 50 receives "anonymized" orders, meaning that the order is stripped of the buy-side customer ID, and identified instead by an internal Order-DD and the sponsoring Broker-ID. Tlie Execution Engine 50 processes new incoming orders by checking to see if there are resident orders that match with the incoming order and executing trades with resident orders ranked using price time priority, as we describe in more detail below. Tks Execution Engine 50 in the preferred embodiment is the book of reconl for recovering the true status about any CIoud9 trades in the event of a failure. The preferred embodiment of the Execution Engine 50 supports Limit orders and orders pged to the NBBO Midpoint. In an alternate embodiment other order types known to those skilled in the art are also supported. For example an alternate execution engine may support price discretion orders that are executable at a limit price agnst market orders, but may automatically increase their price aggression to trigger a match with an incoming limit order up to a predetermined discretion amount, hi another example of an atteraate anbodiment, the system enables traders to eater orders that are to be priced at a later time based on a volume-weighted average price (VWAP) for a forward time interval, such as the VWAP from the time of the trade until end of day, or the VWAP over a specified time interval, such as a half hour interval. The orders in the preferred embodiment can be decribed as limit orders with an optional Peg. All orders are required to have a limit price; if they have a Ped price as well then the order is considered to be priced at the more passive price between the limit and the pegged value. Peed orders are priced on the occasion of a matdi check evait and are then treated the same as a limit order at the more passive of the limit or pegged prices. The pegged value is the NBBO Midpoint plus a Peg Oflfeet amount. Users in the preferred embodiment will use the Peg as a protection on a limit order to ensure that if the market moves swiftly, theh order does not trade at a price that suddenly staads at a very aggressive leve! as compared to the NBBO prices. The GUI 20 suggests tiiis way of using a Peg by representing the feature as "Price protection" and offering the option to "not trade more than cents from the NBBO midpomt price." Because there are both pegged orders and lunit orders in the system, a matdi may also be triggered not on order entry, but as a result of a change in the NBBO prices that would Cause a pegged order's price to cross a contra ordo's limit price. Therefore, in the preferred embodiment of the present invention, a match can be triggered either on entry of a new onJer, or on a diange in fhe NBBO prices that causes a pegged order to cross with other orders. The system preferably implements the following process to identify such matdies. The Execution Engine 50 preferably comprises a component that receives all quote update messages from the data provider 60 and keeps track of the current best bid and best offer, using information caching techniques that are known in the art. It preferably also enables a component of the Execution Engine 50 to subscribe to price updates for a given security, whereafter it will receive a message whenever the best bid or the best offer in the given security dianges. The Execution Engine 50 preferably maintains a list of symbols for which it has one or more Limit Orders on one side and one or more Pged orders on the other side and subscribes to NBBO price changes for all securities in this list. Pegged orders with a limit price count both as pegged orders and as limit orders for the purpose of the present description. For the symbols within this list, Bcecution Engine 50 will process each NBBO price change by pricing all pegged orders in the security and then ranking all orders on the buy side and on the offer side in price tune priority. The system preferably proceeds to identify matches by taking first the top-ranked buy order, and processing trades between this order and sell orders in price time priority. If there are still unfilled sell orders, it proceeds with the second-ranked buy order and again processes trades with the yet-unfilled sell orders in price time priority. The Execution Engme 50 will preferably observe trading halts on the primary market as follows. While a security is halted, ail existing and new orders will be declined. The Execution Engine 50 derives the security-tradmg status fix)m a Data Vendor 60. If the Data Vendor 60 is unable to provide the service or the corresponding communications network 70 fails, the Execution Engme does not process any trades but sunply wwts for the vital services to be re-established. The market data feed in the preferred embodiment contaios the market halt information as well as level 1 and last sale data for purposes of the Analytics Server 160. If a New Order triggered the match, the &tecution Eiine preferably reports the execution pending immediately to Cloud9 OM 130 with the expected total matdi quantity. This report is informational only for use by the Facilitator 140 and does not guarantee that the trade will clear, as we will see below, A trade is preferably processed as a single transaction, as is known in the art, to avoid the risk of the system reporting one side of the trade and not the other in the event of a fiulure. The processing involves tiie following steps. 1. Report the trade to the regulatory organization as required and update the order status in the database. In the preferred embodiment the trade is reported to the Nasdaq ACT system. The matching engme reports to ACT immediately for data dissemination (tape reporting) and for clearing purposes. 2. When the tmde is success&lly locked in for clearing (ACT acknowledges trade reports and assigns an ACT report ID), the Execution Engine 50 preferably sends a finn trade report message to the Order Manager 130. Trades involving pegged orders are preferably reported with the timestamp of the most recent bid and the timestamp of the most recent offer that were used to calculate the NBBO midpoint, as well as the timestamp at which the execution was processed. The Execution Engine 50 processes incoming messages as follows: New Order Processing. New Order is a message fiom the OM 130. It is preferably processed as follows. • Identify matdiing orders, if any, and acknowledge order entry with status to the Order Manager 130. • If the order is not filled but causes a situation where there are limit orders on one side and pegged orders on the other, register for NBBO price lipdates in the security. • Process any trade as described above. Cancel Order. Cancel Order is a request from Cloud9 OM to cancel an open order. It is preferably processed as follows, • Verify whether the Cancel Order request carries a valid order ID that points to an order in the Book 51. • If the order has already been traded, reject the cancel order request message • It the order is not yet traded: Marie the order status as canceled in the Execution Engine database (Book 51). Report the successful cancel status to the Order Manager 130. In the preferred embodiment an order can be modified or canceled at any time. In an alternate embodiment the orders are required to have a minimum time in force of 10 seconds to give other participants a reasonable time to react to events triggered by the Facilitator 140, such as an Active Symbol flag or a Contras present flag. The system preferably also enables the Order Manager 130 to request cancellation of all orders associated with a given sponsoring broker, or all orders altogether. The Execution Engine 50 processes such aggregate cancel order requests by first identifying all affected ordffs and thai processing the cancel order requests by time of entry of the orders. The Execution Enne 50 preferably validates connectivity with the Order Manager 130 using 30~second heartbeats. On loss of connection, the system preferably cancels all orders. NetServer. The system preferably comprises an Application Programming Interface that enables a client GUI 20 to connect to the NetServer 110 through a communications network 80 such as the Internet. NetServer 110 allows customer fi-ont-end applications to access CIoud9 trading services such as the viewing of BPR updates and entry of orders in to the system. If loss of connectivity to a client is detected, Cioud9 preferably attempts to cancel all the client's orders with the Execution Engine SO. In an alternate anbodiment the system allows cliaits to select a configuration wherein the loss of connectivity does not cause cancellation of open orders; clients that elect this alternate configuration then have to call the Help Desk to cancel orders in case of an emergency, NetServer 110 passes trading activity messages (orders, cancel/replace, cancels) to the order manager and passes responses and unsolicited messages bade to the climt. It also stores client GUI configurations, such as the location and size of windows on the sCTeen and the list of securities that the trader wishes to watch. Tliese examples do not constitute an exhaustive list but instead are intended for the purpose of illustration only; otha: configuration parameters are stored to support normal GUI display options as can easily be apprehended by those skilled in the art In a preferred embodiment, the GUI enables traders to view credit alerts generated by the system whai the amount of available credit falls bdow the dollar amount typically required to execute a large block trade. In an alternate embodiment credit alerts are not displayed and the user simply discovers the absence of sufficient credit upon the feifure of an order entry request. In the preferred embodiment of the subjea invention, the GUI 20 enables traders to access a dialog that fiicilitates management of a watch list of securities. The purpose of the watch list is to limit the flow of mfonnation from the sjtem to the symbols that the trader is Interested in. In the preferred embodiment, the system also enables a trader to see whether or not there is someone else with an open GUI cuirently watching a given symbol (this being of interest since it would increase the likelihood of a trade). In an alternate embodiment the GUI 20 enables users to see the number of traders watching a security. The GUI 20 will register for BPR update messages and for receipt of the Active Symbol and Contras Present messages only in watched securities. In the preferred embodiment the API allows a GUI to register for BPR/Active/Contras Present messages in any security, and separately register for quote update messages in a separate list of securities. The preferred GUI 20 only displays NBBO prices for one security at a time, namely the one for which the order entry dialog is open, as depicted in FIG. 2. In an alternate embodiment the GUI subscribes to NBBO updates in all watched securities and these prices are displayed next to tbe symbol in a drop jist To manage the watch list, users are preferably able to add mdividual symbols, or add all symbols that are classified as being in a known group of securities, such as an indxistry group or a seaor. The system preferably also enables users to submit their own groups of securities in the form ofan Excel file and load these groups into the system, at which point they are sunply added to the list of groups of securities that the trader may choose ftom to populate his or her watch list A preferred watch list configuration dialog is depicted in FIG, S. The user is able to add symbols or groups of symbols. The changes are in a pending status until the user saves the changes. Users can also revert to a prior saved watch list or load a group of securities from an Excel spreadsheet which can be, for example, derived fiom their OMS or blotter. The watch list is preferably managed by the client and persisted on the server. In a preferred embodiment Net Server 110 keeps track of which symbols are being "watdied" by two or more GUIs. If a symbol is added to a trader watch list, causing the number of watching clients to reach two, NetServer 110 will broadcast a message to all GUIs that subscribe to active symbol messages to indicate that the symbol is watched. If the number of watching GUIs drops to one, NetServer will broadcast a message to the GUIs that subscribe to active symbol updates to indicate that this symbol is no longer being watched. In an alternate embodiment the message is sent any time a user adds or removes a watch on the symboi and gives the symbol, timestamp, and updated number of parties watching the symbol. The GUI 20 preferably shows all symbols that are both m the trader's watdi list and are either active or have contras present, by highlighting them on the top of the GUI through color-coded rectangles, using the color orange for active symbols and yellow for contras present. In an alternate embodiment the messages indicating that a watched symbol has become active or has contras present is displayed by means of a twnporary popup screen, similar to instant messenger applications. In another alternate embodiment these popup messages stay visible until action is taken, either by clicking to generate an order entry dialog and entering an order in response to the message, or by clicking on a "close" or "minimize" button to either close the popup or reduce it to a bar at the bottom of the screen. TTie trader GUI 20 in the preferred embodim«it has the option of showing or not showing the symbols on their watdi list that are in a "watched" state (i.e., at least one other trader is watching this symbol). If they choose to see these symbols, when they are not "active" or with contras present they will be represented in a third color that is less striking than the other two since the trader's attention is not immediately required; in the preferred embodiment the GUI 20 presents watched symbols in a gray rectangle next to the active and contras present rectangles. The NetServer 110 preferably exposes an API for remote clients to access the system's sCTvices. The API preferably enables the user to open a session though basic authenticating (usemame password) prior to using any trading services. In an alternate embodiment the client authenticates itself wi a certificate. The exchange of user credfflitials is preferably done via a SSL session, which exchanges encryption keys using the Net Server public key. Once the user is authenticated the session maps to the users identity. The NetServer API preferably exposes the following functions, * Get Sector List: Eadh firm can optionally have securities arranged by sector or other security grouping. The security groupings can be configured per client film. The API returns the list of securities associated with the cUait's firm and group name. • Get Security List: The system has a list of configured securities. TTiis function returns that Jist Each dement of the list is the Symbol, Company Name, and Sector. ' Set & Get Trader Defaults: The system enables a client to store default parameters. The following list is not intended to be exhaustive, but contains some of the trader defaults that are of particular interest; others will be apparent to those skilled in the art. De&uit Placed amount. The order entry dialog will populate the quantity field by default with either the minimum quantity for this symbol (Large Block Quantity) or with the trader's total ticket size, if this traderuses tickets. Default Price. The order entry dialog will populate the price field with a limit price chosen via one of the following rules: Aggressive end of the NBBO, passive end of the NBBO, NBBO Midpoint, Aggressive end of the BPR, Passive end of the BPR, or BPR Midpomt. Default TIF. The order entry dialog will populate the time in force field with this number of seconds. Default Watch List Management (Security, Sector). All watch list managanent is handled by the fiont end and persisted on the server. ° Default Peg and offeet. Tlie order entry dialog will select or not select the Peg option by default and if selected, it will add this number of cents as peg offset for buy orders, or subtract this number of cents to the midpoint peg for sell orders. Ticketed Orders Only. This option determines whether a trader wishffi to check orders against the size reserved by means of a ticket If this option is not selected, the trader will be able to enter orders even when there is no corresponding ticket. Set & Get Sector Watch List: Save and retrieve a list of securities that comprises a sector or other group. Set & Get Symbol Watch List; Save and retrieve the list of securities that are in a given watch list Set & Get GUI Defaults: Ticket Management Grid: option to start the application with the window that displays the list of tickets open or hidden. Ticker Management Grid definition: font, field order, col size, col type, headers and field=col mappings. Active Symbol list (open or close). Option to start the application with the window that shows the list of active symbols open or closed. Color Changes (Yes, No): The GUI preferably allows users to select this option to highlight any field of an order in the order entry dialog when the order has already been submitted to the system and this field has since been edited. When a field has been edited the user can press a "replace" key to send the new value to the system via a Cancel/Replace order message. Highlighting the fields that have been edited assists the user in managing an order. Get Heartbeat Interval. Number of seconds for the periodic heartbeats to determine whether the connection between the GUI 20 and Net Server 110 is operational. The system preferably imposes a lower bound on this pafameter to avoid overloadingNetServer 110 with connection management duties. In an alternate embodimrat the parameter can only be set on the server and is not exposed via the API. Get Tickets: gets the user's tickets. Within the trading day, the user cm only have one ticket per symbol/side. If the ticket is cancelled, the size decrements to the unexecuted size. New tickets with the same symbol and side replace the entry in the list. Get Orders: Gets a list of all orders. Get Order Status. Returns the order details for the ClientOrderlD. Submit New Order: Queues a new order into the system. Asynchronous call that will be followed up with OnOrderUpdate or OnOrderReject event(s). Replace Order This function queues a replace order request. The caller assumes the order state to be in "Pending Replace" upon retum. Asynchronous call that will be followed up with OnOrderUpdate or OnOrderReject event(s). Replaced Order details are the same as New order details + OrigClientOrder ID field. The quantity is the total quantity including filled size. Cancel Order: Tliis fiinction queues a ranceled order request. The caller assumes the order state to be "Pending Cancel" upon retum. Asynchronous call that will be followed up with OnOrderUpdate or OnOrderCancelled event(s). • Execute Trade. This function requests tiiat this trade be reported to the sponsoring farokerto setup allocation process. This would be done automatically at the end of the trading day. " Get Trades for (ClientOrderlD, Symbol, or ALL) Request/response to retrieve trade details. There are preferably three versions of this request. Get trades for an order, for all orders in a security, or for all trades for this GUI client. The following are API events - i.e., messages produced by the system and pushed to the GUI 20 through the API • OnTdcetUpdate: this event is raised on new, replaced, and cancelJed ticket events. • OnOrderUpdate: Event raised on order state changes. This includes order management responses to the requests above or through other channels. • OnOrderReject: Event is raised in response to a new/replace order request entered within this session. " OnCancelReject: Event is raised when a cancel request is rejected, • OnTrade event is raised when there is a trade within one of the traders orders entered through this API or another channel. The preferred embodiment of the NetServer 110 also provides an interface for subscribing to market data per security. Fields include: (NBBO bid, ask and TapeTimeStamp), (BPR Low and High), Block Tape (Last and total volume), Active Symbol, Contras Present, Security State (open, closed, halted, etc), and limestarap. The functions exposed in the preferred embodiment of this interface are: • Subscribe Request/Response; This function preferably adds a list of securities provided in the request message to the client's subSCTiption list The request returns a snapshot of that entry's current data for the requested fields. Subsequent changes to the watched fields will raise an update event (OnUpdate) with the changes. • Unsubscribe Rajuest/Response: Removes the subscription request. • OnUpdate Event: Notifies the subscriber of fietd changes within a Subscription entry, ■ Start Active Symbol Feed Request/Response: Returns a list of all symbols where the active symbol flag is set. Subsequent to the call, all state changes in active symbol flag changes will result in OnActiveSymChange(Symbol, On/Off). • Stop Active Symbol Feed Request/Response: terminates above event notifications. • OnActiveSymChange event notifies the subscriber of active symbol flag dianges Research Data Storage. The system preferably stores mfonnation that allows operators and research staff to evaluate how indicators of trader activity in a symbol correlate to fill rates when users enter orders. Tliis information plays an important role in steering usage of the systan toward workflows that lead to greater success, through marketir media and visits to the traders' stations. For example, if the fill rate following entry of an order in a security that is not watched is extremely low or null, traders may be advised to focus their trading on the system in securities that are watched by other traders. The system preferably also can be reconfigured to modify the rules of when flags are sent out, to improve fill rates in the system. For example if it is detwmined that fill rates are substantially higher when there are three or more traders watching the security than two, then the system can be reconfigured to send out the watched symbol message only when the total number of GUI users is three or more, rather than two or more. Tie above examples are intended for the purpose of illustration and not to provide an exhaustive list; other optimizations based on modifications of parameters given above will be apparent to those skilled in the art. The trader activity measures are available on the help desk, and exported to a permanent data repository at the end of each day for offline analysis with the purpose of fmding correlations between possible system configuration settings and higher fill rates. The measures of interest are: • Order activity (entry, BPR aggrasion changes; cancel/expire) Fills • Tickets • Number of traders that can see the symbol active flag (on watch list, as a symbol, or as part of a sector) • Number of traders that can see the BPR updates (subscribers) • Number of traders that can see the quote updates (subscribers) The data tables below (A-D) are stored in standard comma-delimited (CSV) format. Analytics Server. The Analytics Server 160 in the preferred embodiment keeps trade of the National Best Bid and Offer prices by listening to a vendor quote feed 60 and updating said NBBO prices when the quote that made the best price is either canceled or improved by a new quote with a better price. The current NBBO prices are preferably stored in a cache using methods well known to those skilled in the art. In addition to the NBBO prices, Analytics Server 160 calculates a Block Bid and a Block Offer based upon recent quote and trade data. The Analytics Server's main fijoctJoD is to produce the following messages." • NBBO price changes. " Block Price Range updates. TTie analytics server will store last sale data and all changes in the inside bid or offer prices throughout the day, in all supported securities. The data that needs to be stored for analjis is described in the Tables E and F. The Analytics Server 160 delivers quote update messages to all connected clients whenever there is a change in the inside market best bid price or best offer price. The update message carries the new bid price or new offer price and timestamp associated with the appearance or removal of the quote that dianged this bid or offer. The Analytics Server 160 preferably also updates the BPR every 60 seconds, and sends a BPR Update message to the Order Manager 130 for the purpose of determining the price aggression of its orders and to Net Server 110 for broadcasting to GUI Clients 20. The preferred embodiment comprises a replaceable module that is responsible for calculating the BPR. The BPR can be calculated using methods known to those skilled in the art, such as for example by taking a Block Bid equal to the National Best Bid price minus 5 cents, and a Block Offer equal to the National Best Offer plus five cents. In another example, the amount of cents to be added (subtracted) to (ft'om) the NBBO prices is set in proportion to the historical volatility of the stock, so that for very volatile symbols such as technology issues the variation might be greater than 5 cents, while it would be less for more steady stocks such as blue chip utilities. Tlie algorithm used in the preferred embodiment of the present invention considers the highest and lowest prices that have traded in the past 60 seconds, H and L respectively, as well as the current NBBO Midpoint price M, and calculates the Blodt Bid and Blodc Offer through the following steps [FIG. 9]: ■ Step 910. Determine the prices H, L and M for a given security as defined above. • Step 920. Looking at last sale data back to tlie timestamp of the last BPR update, find the trade that was printed furthest from the current NBBO midpoint. Let X draiote the absolute price difference between this trade price and the current NBBO midpoint. Thus, X is the greater of H-M and M-L. Step 930. Let Z = Y + (MaxBlockSpread/2 - Y) * (1 + 1 / (1 + exp(-BETA*X)), where Y is a parameter set per security that sets a lower bound on the spread between the Block Bid and Best Offer, MaxBlodcSpread is the upper bound on said spread, and BETA is a parameter, whid: can be set to 10.0 for example. • Step 940. Calculate the Block Bid to be equal to M + Z, and Block Offer equal to M - Z. • Step 950. Issue BPR Update messages to the Order Manager 130 and NetServerllO. Credit Management. The system preferably enables sponsoring brokers to set credit limits on their customers' accounts. In the preferred embodiment, credit checkmg is based on a gross dollar limit, counting the total value of shares bought plus shares sold. Orders will be vaHdated for credit based upon the greater of the top of the BPR or the order's iimif price, and credit will be adjusted based upon the actual credit amount consumed when the order is completely filled, expired, or canceled (on partial fills). • Brokers will be given a web account to manage credit; they can view the . level of credit currently consumed by a client, but only binned as follows: less than 25%; 25-50%; 50-75%; more than 75%. * Brokers must be able to increase today's credit with immediate effect; or they can increase or reduce the amount of credit that gets refreshed at the start of each trading day. A broker can also suspend credit; this has eft both intraday and for following days. If day's credit is suspended, the system will cancel all open orders. ' The broker can later un-suspend credit for a client Suspensions have permanent effect and persist into the next day. Credit is refreshed at the end of the day. • The broker will be able to set up a credit alert threshold as a percaitage of the total (default =* 75%); additionally, the system will geaate an alert whenever the available credit falls below a system-configured dollar amount MinCreditAmount that is too low to utilize the block system (for example, $2million). Credit alerts will be delivered to Help Desk personnel and pushed to the client GUI. Help Desk. The systwn preferably provides a Help desk interface to be used by personnel that handles calls fiom customers. This user interfece can be a web browser mterface, witli access secured through Public Key Encryption and certificate-based authmtication as well as a usemame and password pair. The interface preferably enables the user to view detailed information about the life of an order. The user can pull up the list of orders with a given symbol and client ID to locate the order that the customer is calling about. The help desk interfece enables its usw to click on an order of interest to view the following messages in chronological fashion. * Ticket, if applicable. • Order Entry request(s). * Order Entry Response: rejected (with reason code); execution pending; expired; open. Fills. Contras present flag(s) sent against an order. Modify Order Requests, and responses (accepted, rected). Cancel Request and response (accepted, rejected). Expiration. In addition, the help desk operator is able to see the NBBO inside prices, timestamp of tbe most recent quote seen on the bid and offer sides, respectively, BPR prices and timestamp, on each significant evMit in the life of an order (entry, cancel, execution). This is of use for answering questions such as why the symbol did or did not change to an active state when the order was entered, «c. In an alternate embodiment the help desk operator carmot view symbol or side of any order, to reduce the security risk associated with traders being aware of the trading interests of institutional customers. In this embodiment, the help desk operator identifies a caller's order by pulling up the list of orders associated with the trader, with the time of entry and size of each order as well as the ClientOrderlD which is also visible on the trader's GUI 20. The help desk operator woiks with the trader to identify which order he or she is inquiring about, then proceeds as above by clicking on the order to view the corresponding activity trail. In the preferred embodiment the Help Desk also enables its operator to select a client firm from a drop list, and view their configurations pertinent to trading logic. The list below gives the more unportant user configurations and usage statistics for the help desk operato , the list is not intended to be exhaustive, others will be easily apprehended by those skilled in the art. Credit limit and credit consumed. Activity by trader: number of orders, traded shares, traded dollars. List of trader, select one to view and edit trader options. Watdi list management (add/remove symbol, add/remove mdustry group). Show / don't show active symbols bar. Show / don't show drop list of tickets. Default ticket attributes. The preferred embodiment also enables brokers to access a web page using the same authentication and security model as for Help Desk operators. Brokers will use this page for credit management purposes. In order to maintain the client's confidentiality, the interfece preferably limits the information displayed to the broker in such a way as to not reveal the details of the client's trades. In this embodiment, the broker can view the amount of wedit consumed by a client only as a percentage of the total credit limit; for example, this can be binned in four intervals: less than 25%, 25-50%, 50-75%, and greater than 75%. The sponsoring broker additionally is enabled to modify the client's credit through the following functions. Add intraday credit. ° Add credit today and also increase the refresh amount for following daj. ° Lower refresh amoimt for following days. Suspend credit. Uosuspend credit. In an alternate embodiment the trades are reported to the sponsoring broker immediately and the sponsoring brokers are able to view executed trades as well as the precise amount of credit consumed. In another alternate embodiment the sponsoring broker can also view pre-trade activity. System configurations interface. The system preferably enables an operator to modify the configurations, and add/remove a client or sponsoring broker, during an off-hours system maintenance schedule. The system configurations interface enables the system operator to add a client and set required configurations attributes to enable the user to trade. New client firms must choose a sponsoring broker from a list of available brokers. The sponsoring broker is preferably unique for all traders within the same fimi. In an alternate embodiment the same firm can use a plurality of sponsoring brokers, and the GUI 20 lets the trader select a sponsoring broker on order entry. The new firm can optionally set up FIX communications directly to FIXServer 120, or through the sponsoring broker 95. In the latter case the sponsoring broker forwards FIX messages from the client to FIXServer 120 on its behalf A firm may selea two modes of FIX connectivity depending on their workflow requiranents; FIX channel can be set up to receive Executions only, or to receive Executions and Order updates. They can be set up to enter tickets and chedc GUI-entered orders against the size allocated through a ticket, or to work without the use of tickets. When a client finn is added to the system, the Help Desk personnel will call the broker to set up any credit limits for the customer's account, and agree to a client-ID to be used fcrendofdayreporting. Newusera are preferably required to configure GUI/API access to be able to trade on the system. In an alternate embodiment order entry is also supported via the FIX interface, and the user is not required to use the GUI 20 or API access. The system configurations interface preferably also enables the systan operator to add a sponsoring broker to the list of supported brokers. In an alternate embodiment the system is operated by a smgle sponsoring broker and cannot be accessed through a third party broker relationship. In configuring a sponsoring broker, the operator will create a user account for tiie sponsoring broker to manage credit for their clients' accounts as described above. Broker contact information such as a telephone number, FAX, and email must be entered and stored in the system database 150; Help Desk personnel will utilize this data to contact the broker for credit issues. To complete the process of setting up a new sponsoring broker the following steps preferably are executed. • Brokers must set up a FIX connection to receive drop copies of all executions. If the broker is also used as a service bureau for the client's OMS and sends Tickets on behalf of the clients, then this same connection will be used also for FIX connectivity to the client. • Set up end-of-day file transfers from the system containing the details of each trade, including the identity of the buy-side customer for each trade. • Optionally, set up end-of-day file transfers to the broko's clearing firm at end of day with the sell-side trade details. The operator is forther enabled to add or remove securities from the list of securities supported for trading. The securities are preferably associated with parameter values specific to the present invention, such as the Large Block Quantity and parameters utilized by the Analytics Server 160 to calculate the Block Price Range. An example of a set of required fields for a security is given in the table below. System operations. TTie system in the preferred embodiment preferably comprises an operations console used to provide centralized management of the servers, network hardware and trading. Operations software to support required fimctions is commercially available in versions that minimally provide the following features: Real-time Display of all server statistics Problem Summary Problem Resolution Feedback Remote Management / Monitoring Intelligent Notification System (Alerts) Alert Display, Acknowledgement, and Annotation Audible Alanu Logging Online Help actions: The operator console allows the system operators to execute the following remote System Startup System Restart System Shutdown System Component Restart System Component Shutdown System Component Startup System Backup Database Backup Inner Day Startup/Shutdown • End-of-Day Reset • Execute End-of-Day Batch Processes The system additionally enables the operations staff"to generate the Daily Usage Reports at the end-of-day and extract from these the necessary information for billing, research, clearing, and OATS reporting. The operations staff preferably also uses a system operations tool to create a clearing summary, which contains a list of trades with the broker JDs for both sides of the trade. Trading summaries pertinent to a sponsoring broker and its clearing firm contain all trades where at least the given broker sponsored one of the sides. The system preferably also generates billing reports containing a list of all fills involving a sponsoring broker for sending to the broker. The fills should match intra-day execution reports one-to-one. The end-of-day report additionally contains the client ED, which was masked in the intraday reports. Ifbothlegsof a trade were sponsored by the same broker, there will be two fill reports in the summary file. The preferred embodunent saves two copies of the data described below to a removable medium for archiving and analysis. Wherever possible, the logging of events includes a date-time stamp with millisecond accuracy. • System and Application Logs: The application and system logs on each machine are labeled and saved. The Log is preferably reset before begin-of-day. • Configuration Audit: The details of a begin-of-day and end-of-day audit of the firmware configuration are recorded (NT Registry, application Config, etc.). • ACT Messages: There shall be a separate log contaming the details for all ACT messages. This log is preferably rraet before begin-of-day, ' Orders & Trades: The order and trade tables are copied to removable storage at end-of-day. A separate log for the details on order state transitions is maintained. Log and tables are preferably reset before begin-of-day. • Operation Activities: The operator activities affecting the behavior of the system are logged. The logging will be time stamped and include the identity of the operator. This log is preferably reset before bin-of-day. All System Activity Logs and Daily Reports are preferably moved to permanent storage at the end of each trading day. System Activity Logs contain enough information to reconstruct the trading logic and pricing for each order and trade event, including time stamps required to resolve pricing disputes. The system is preferably configured to automatically recover services in the event of an accidental failure, using processes based on the considerations below. ' Loss of GUI. Tlie GUI 20 is a required channel for Cloud9 Services. If Net Server 110 detects a loss of connectivity with the client, in the preferred embodiment all orders are automatically cancelled and the sjtem generates an alert to help desk personnel with the phone number of any trader with canceled orders. The GUI 20 shows the order status as "cancel pending". On reconnecting, the order status is updated to "canceled". Traders can pull up a list of orders that were canceled as a result of the connection failure and re¬activate these orders; the system will treat them as new orders and process them accordingly. In an alternate embodiment, users are enabled to select a configuration where their orders would not be canceled on loss of connectivity with the GUI; in this embodiment the user will call the Help Desk to cancel orders when the connection caimot be recovered. • Loss of FIX. FIX connectivity loss preferably does not cause the cancellation of orders, because FIX is not the channel through which executable orders are managed. The operator must be alerted upon loss of a FIX connection. Upon reconnecting, the FIX client will re-synchronize with the server in accordance with the FIX protocol requirements. In an alternate embodiment FIX can be used to enter executable orders in the system and said system supports a configuration enabling users to request fliat their orders be automatically canceled upon loss of the FIX connection. • Analytics server. If the Analytics Server 160 is out of service, the system in the preferred embodiment will continue to fiinction but without the benefit of fimctions normally implemented by the Facilitator such as new active symbol messages or Contras Present messages. The GUI 20 will also fiiil to receive quote updates and BPR update messages. If the Analytics Server 160 cannot be brought back on line swiftly, system operators have instmctions to place the system in an off-line state where it rejects any further order raitry, but does not cancel existing orders. These solutions are described as an example of how a trading system such as that in the present invention addresses reliability issues, but other solutions will easily be apprehended by those skilled in the art. For example, an alternate approach is to consider the Analytics Server 160 as a vital service and automatically take the system off line and cancel all orders upon loss of service. • NetServer, Order Manager and Facilitator. These services are closely related to vital firactions of the trading systan. Failure of these services is therefore considered fatal in the preferred embodiment; the operator will attempt to cancel all ordeis in the execution engine or these will be canceled automatically when loss of connectivity is detected. TTie Cloud9 systwn automatically goes in an off-line state upon detecting failure of any of these sauces. The integrity of the system is constantly monitored by a System Management Console, whidi runs continuously on an independent system. I at any time during the trading day, the integrity of the system becomes uncertain, the system automatically switches to aa off-line mode. At this point all ordeis are cancelled. All iurther orders are declined until the system returns to a fully operational state. System-imposed symbol active flag In an alternate embodiment, the system automatically sets all watched securities to an active state at announced times and maintains these symbols active for a 15-minute wmdow. For example, the sjtem may set all watched symbols to an active state at 11:50 and hold them aaive until 12:05. This alternate embodiment with an active window helps focus the attention of traders around a given time, and allows traders to participate who would not feel comfortable with placing the first order that causes the symbol to become active. It also helps traders manage the space on their workstation screens, by limiting the amount of time during wMdi they are expected to dedicate space to the GUI 20. Finally, by utilizing a call window that has already become known to traders as associated with block crosses thanks to the marketing presence of the POSIT® matching system, this alternate embodiment can leverage the fact that tradffl are aheady used to seeking a block crossing opportunity at this time and focus liquidity in CloudP for symbols where a plurality of traders are watching the GUI 20. Upon termination of the active window, a symbol's activity state reverts back to what it would have been based upon the description of the preferred embodiment Thus, if there are open or recently canceled orders in the system in a given symbol upon termination of said active window then said symbol will remain active; other symbols that have not seen order activity will revert to a "watched" status. As long as a symbol remains in an active status, traders can continually place new orders without this act causing the symbol to change activity state. For very liquid symbols, this can potentially lead to a symbol staying in an active state through the end of the trading day. While in this mode, the system will focus traders in price by means of the contras present flags, but there is no further need to focus in time. Watched symbol flag In an alternate embodiment, the system is as in the preferred embodiment but additionally indicates to users when a symbol is being watched by at least two traders. If the symbol is on their own watch list, they will see when there is at least one other trader interested in the symbol. For example, the watched symbols that are on the trader's watch list appear in a gray-colored box 1020 on top of the user interface, similarly to the orange-colored active symbol flags 1005 and yellow-colored contras present flags 1010, as shown in FIG. 10. The purpose is to provide the trader with conditioning information that increases the likelihood of a fill. The fill rate would he expected to be higher when the symbol is watched since at least one trader will see the symbol become active upon an order entry event. Of course a trader who has a watched symbol on his or her own watch list can always find out if there are one or two other traders watching in addition to him, simply by removing the symbol fi-om his watch list. To make the system more user-friendly, an altMnate embodiment also shows the trader in this case whether there is one other trader watdiing, or two other traders. Itis can be done by means of an asterix next to the watched symbol name to indicate whsa there is only one other trader watching. In an alternate embodiment the system shows the user the number of traders that have this symbol on their watch at any time. In counting tiie number of traders that are watching a security, the system preferably looks for traders that are in fact able to see an active symbol flag. In one embodiment this is implemented by counting only GUIs that are not minimized and are not detectably hidden behind another application. In another embodiment the GUI cannot be minimized without leaving a residual alert mechanism. For example the minimized bar in the tray at the bottom of the trader's screen will flash with the orange or yellow color when a new active symbol or contras present message is received, and show the symbol associated with the most recent such message. In yet another alternate embodiment, the number of traders that watch this symbol is not displayed, but the trader can learn the number of parties watching upon entering an order. In this ahemate embodiment where the trader does not know if a symbol is watched xmti! submitting an order, the trader can also check an "auto-cancel" box to request that his order is automatically canceled if no other trader is watching the symbol, since no one in that case would be able to see the active symbol flag and fill rates would be exceedingly low. In another alternate embodiment the trader is enabled to selea the minimum number of traders that should be watchmg, and if the actual number of watching traders is less have the order canceled automatically. Alternate anbodiment with system-generated calls A well-known problem with large block crossing systems is the low probability that a large buyer and large seller enter orders at the same time. Much of the present invention is directed to solving this problem after the first trader has entered an order, but it does not in itself provide guidance as to when this first participant should be most encouraged to place an order. In an alternate embodiment, the sjtem operates as described above and also generates system-generated "call" events that focus traders' interest at times when there is indication of both buy and sell interest in the security. In another alternate anbodiment the active flags are not sent and the call is the only time focusing event. The call is preferably schaiuled via an algorithm that determines when the likelihood of a fill is highest. It preferably does not utilize unbalanced trade information (such as revealing the existence of buy interest in absence of any sellH'), so as to avoid the perception of information leaks whereby traders would feel that the orders they have placed in the system are causing unwanted information events beyond their direct control. Thepurposeofthecallistoattractorderentry from traders when the likelihood of a fill is greatest. For example, if the average fill rate for a first trader placing an order in a given security were 5%, the fill rate when a call is open in the security might be 10% or 15%; in contrast the fill rate when there is no call would be lower than 5%; with an overall average of 5% for a trader that places ordere at random tunes independentiy of any calls. Minimally, the system will generate calls to focus order entry in a security when traders are missing each other on both sides. Other opportunities for generating calls are outlined below. To avoid information leak issues, these calls will be reported together with the BPR updates rather than together with an order Mitry event. In an alternate embodiment they are issued on the half hour, to fiirther increase the traders' focus. The algorithms to generate calls are preferably based upon the following principles. • Not triggered on order entry. Calls will be timed with BPR updates. Tlus eliminates the fear of information leaks that would occur if a trader were to see his own order immediately cause a call. • 1 Minute Minimum lifespan. Once a call has been initiated, it will stay open for a sufficient time to let traders react and respond to the call. • Two-sided. Calls will be issued only if there is evidence of block interest on both sides. This reduces potential gaming. • Not necessarily an order. Calls may be generated whoi there is no active order in the system. This again reduces gaming problems. • Undisclosed rules. Again, reduces gaming issues. Also allows the algorithm to be updated (within the context of the above rules), and tune it to maintain a reasonable frequency of calls. The following data is preferably available for the purpose of deciding when calls should be generated. " Tickets with remaining quantity, " Open orders. " Expired orders. • Trades. • Block prints on the tape. ■ Repeated prints on the tape with unusually high volume. Rules-based mgthod for issuing pallg A first alternate embodiment utilizes a niles-based approach for deciding when the call should be issued. The Activity Evaluation process is executed together with the BPR Update process, for each symbol. If die isn't already a call, the rales givea below are preferably applied to decide if a symbol should be called depending on activity in the system (orders, trades, etc.) and on the market. If any of the Boolean rules are true, the symbol will be called. If a symbol is already called, has been called for at least seconds (configurable global parameter), and there is no contras present flag in the security, the system preferably removes the call. Rules that lead to a call are listed in the bullet points below. An order was entered and has expired. Later, another order is entered on the contra side. At the next BPR refresh time, che An Order was entered and has expired. Later, a block print is detected on the tape (greater than 10000 shares or $250,000, agn these are all configurable parameters), and the call has not been posted in the last 90 minutes in this symbol (configurable ActiveMaxDeiay2). Discretionary: a help desk operator types the symbol and elides on a button. An order was entered and a ticket exists on the contra side. An order was entered in a symbol that has traded within the last minutes. A residmt order is priced passively and a new order has arrived withmtheBPR. A resident sell order is within the BPR and repeated prints are detected on the tape indicating aggrate buys at the offer for a block size. The above list is intended for illustration and is not intended to be exhaustive. Other rules that could be used to identify when diere is a greater likelihood of a trade will easily be apprehended by those skilled in the art. Score function method for issuing calls. Another alternate embodiment with calls utilizes a score function to determine when to generate a call, as described next The Call Evaluation process is executed together with the BPR Update process, per symbol. If there is not currently a call in this symbol, the system calculates both aU activity score and threshold for the symbol, as described below. If the score exceeds the threshold, the system preferably generates a call for this symbol. If a symbol is already called, has been called for at least seconds (configurable global parameter), and there is no contras present flag in the security, the system preferably removes the call. The system preferably calculates the score function via two steps. Step 1: compute the Buyside Interest and Sellside Interest in the security by adding weights associated with the following conditions. • OpenOrderCondition. Buy (sell) order currently open and priced within the BPR. • PassiveOrderCondition. Buy (sell) order currently open but priced more passive than the BPR. • E3q)iredOrderConditioE. Buy (sell) order expired since symbol was last active. • TicketCondition. Buy (sell) ticket with remaining quantity. • BlockTapeCondition. Large Block Fill on Cloud9 since symbol was last active. • MaricetPrintCondition. Block print on the tape for at least shares or , printed above (below) the midpoint, having taken place since the last BPR update. • RandomRefreshCondition. Repeated prints above (below) the midpoint since the last BPR update with toUl quantity above shares or , at a rate that exceeds times the average volume for the security. " WatchListCount. Number of traders that have this symbol on their watch lists. The table below gives an example of a reasonable configuration for the weights. Step 2: Compute the symbol activity score as the produa of the Buyside Interest times the SellSide Interest. The system preferably calculates the threshold as follows. The threshold is equal to zero if the symbol has never bem cail. If the symbol has previously been called, the tiireshold for this symbol will be equal to an exponential function of the time since the last call expired in this symbol'. Threshold = MaxThreshold " EXP(-Beta * HmeDelay) The parameters MaxThreshold and Beta are preferably configurable per security. Alternate embodiment with targeted active symbol messages In an alternate embodiment, the system operates as described above except that the active symbol message is not sent to all traders who have placed the symbol on their watdi list; instead, the flag is only sent to traders that have provided certified trading interest information indicating a likely interest in either buying or selling a large block of this security. For example, such certified trading interest information may be a dtxip copy of execution reports that brokers issue following each trad and the selection critaia maybe that the firm has bought net (or sold net) at least 10,000 shares of stock within the last 30 minutes. Other examples are described extensively in the following co-pending U.S. patent applications: 09/585,049, filed June 1, 2000; 09/750,768, filed December 29, 2000; and 09/870,845, filed May 31, 2001, the entire contents of each of which are incorporated herein by reference. Alternate embodiment with a system-proposed match price In the preferred embodiment described above, the traders themselves set price conditions on their order that will determine the execution price with no further intervention on their part. In their current workflow. Institutional traders are accustomed to sending market orders to brokers and accepting or rejecting prices that the broker may propose for executing a large blodc. In this sense, one could say that the Institutional trader is a price "taker," not a price "maker." Electronic trading systems suffer from the problem that the first party entering an order is writing the price that others may ele(a to take or not take; thus leading to an adverse selection problem where their prices will tend to get taken when they committed a mistake and offered a too aggressive price. Attempts to reverse price priority by giving the standing order improvement to the second order's limit, or having the two meet halfway if the prices cross, don't work well because such rules only cause the second order to start with a passive order and then progressively increase price aggression until a match occurs at the standing order's limit. The purpose of the alternate embodiment described below is to solve this problem by having the system generate a proposed match price that is not sensitive to the prices of existing orders in the system. In tins alternate embodiment of the present mvention, the system performs as desaibed above for limit orders and pegged orders, but additionally enables traders to rata" a new order type that we will refer to herein as a "Market Order". A market order may optionally also have an absolute limit price, which in turn may be fixed or pged to a market indicator such as the midpoint of the National Best Bid of Offer. An incoming market order will be handled as follows. Upon receiving a maricet order, if there is no contra order currently residing within the system, it simply stores the market order. If there is a resident contra order, the systan will randomly determine a possible trade price by choosing a price within the BPR with a statistical distribution, such as, for example, a flat distribution (the flat distribution being that wherein all price points within the BPR are equally likely to be sel ected). In another example, the proposed trade price is sensitive to orders in the system, and the users submit to fair arbitration by allowing the system to use order information to determine a fair price. For example, such an embodiment can generate a random price based on a distribution that is not flat but piecewise flat, with two flat segments above and below the BPR Midpoint, .to give greater probabilities to the prices below the BPR midpoint when *ere are more sell orders than buy orders in the system, or vice versa if there are more buyers. For example, the probability of prices below the BPR Midpoint can be taken to be equal to P(Below Midpoint) = 1 / (1 + EXP(Gamma X)), where Gamma is a positive nimiber such as 2, and X is the relative difference between buyer shares and seller shares priced within the BPR: X = (BuyQuantity - SellQuantity) / (BuyQuantity + SellQuantity). Thus, X -> 1 if there is more buy interest than sell interest, and X - -1 in the opposite case. TTiese examples of methods to gaierate prices ate intended for the purpose of iUustratioa and not as an scfaaustive list; other mechanisms to generate a price can easily be apprehended by those skilled in the art. Having determined a possible trade price, the system will proceed to determine the feasibility of a trade at this price according to the following logic. 1. If the randomly generated price is within the limits of both orders, the trade is executed at this price with no fiirther input needed fiiim either player. 2. If neither order is able to accept the proposed randomly-generated price, this price is proposed to both traders, who are 1iien given a time window in whidi to accept or decline the price. For example, the time window can be synchronous with the BPR refresh time, or a minimum of 15 seconds, whichever is longer. At the expiration of this time window a new randomly-generated price is produced together with a new BPR, and this pnDcess continues as long as there are orders on both sides within the BPR. 3. If one order's limit price is sufficientiy aggressive to accept the proposed trade price but the other is not, the price of the order that was not sufficiently aggressive to trade will be proposed to the more aggressive trader, and the randomly-generated price will be shown to the less aggressive trader. Ndther trader will know whether the price they are looking at is the other trader's limit or the randomly generated system price. At the expiration of a time window, a new price is generated and the process continues as described as long as there are orders on both sides within the BPR. In an alternate embodiment, when only one trader's limit overlaps the system-generated price, this traderhas the option of accepting the other's limit or of proposing a counter offer to the less aggressive trader. In this embodiment, the less aggressive trader does not see a price until the first trader has taken aaion, or a 30-second time window expires. The more aggressive trader may cancel his order within this 30-second window without having at any point revealed its existence to the other trader. If the first trader takes no action, the system automatically shows the second and less aggressive trader the system-generated price after 30 seconds. TTiis delay overrides the immediate display of a contras pres«it flag to resident orders of the preferred embodiment. Competition present flag In an alternate embodiment of the present invention, the system performs as in the preferred embodiment but in addition provides a mechanism for traders to know when there is competition to buy (sell) the security, enticiag than to increase their price aggression and tberdiy also fecilitating the process of discovering a fair price for a trade. In a first alternate embodiment with the competition present flag, the system simply lets each trader with an order in the system that is already seeing a contra present flag know wheliier there are oilier orders on the same side. In a second alternate embodiment with the competition present flag, the system performs as in the first alt&nate embodiment just described, but additionaUy the trader with the lower matdung priority based price and time priority ranking is enabled to see that the other trader has a higher matching priority; this trader is then induced to increase the price aggression on his order. If this becomes the new highest matching priority order, the other trader will in turn be allowed to see that his price is no longer the best. This process continues as long as there are multiple orders on the same side in a context and one or more contras are present. Working orders and integration with third party execution svstems In an alternate embodiment, the system operates as described above but additionally users are enabled to request that their orders be worked on the market while waiting for a match within CIoud9. Users that request a working order option are preferably required to enter an order quantity that is greater than the minimum block quantity. The additional size will be dispatched to be worked by an automated process that automatically places small slices on the regular market, in the manner known in the art as "random refresh" orders. Several sudi automated processes are available on the market today; some that are known in the art are random refresh algorithms that place a small quantity of shares at the best price on the market, and whenever said small quantity is exhausted, generates a new small order to be placed at the new best price. The refreshed quantity is chosen at random between a minimum and maximum size, for example between 500 and 900 shares. In another example the order is refreshed after a random delay of a few seconds. Other more sophisticated algorithms are commercially available to automate the execution of an order through its reduction to small pieces that are executed independently, such as ITG's Quantex. Other destinations, such as NyFIX Millennium, enable completely hidden orders to intercept order flow that was on its way to a third maricet destination. The alternate embodiment described herein manages the order across one or more working order destinations in addition to Cloud9. In one version of this embodiment the system splits the order into several chunks, one for the Large Block Quantity is placed on Cloud9; others for a system-configurable quantity (for example, 5000 shares) are routed to external destinations that have the capability of working the order. Whenever a destination has completed its block, the system determines if there is residual quantity besides that which is already placed on execution systems, and if so sends the lesser of the system-configurable quantity or the available residue to the destination that has completed its block. When one external destination has completed a chunk and no residual quantity remains, but there are still incomplete blocks on other market destinsitions, the incomplete blodc that has spent the longest amount of time without receiving a fill is canceled and re-routed to the destination that reports having completed its block. When there are no further mcomplete blocks anywhere except for the Large Block Quantity order on Cloud9; the system sends a message to the user's GUI 20 to notify that the working order component has been completed. The tradCT may then place additional shares to resume working the order, or convert the Large Block Quantity itself into a working order. In this alternate embodiment, the system preferably removes the order from consideration in assigning flags when the quantity remainmg in Cloud9 has dropped below the Large Block Quantity. In another alternate embodiment, the remaining size in Cloud9 can fall below the Large Block Quantity and maintain the order's effect to keep the active symbol flag. In this embodiment, a second party that has also selected the working order option may receive a partial fill against this CIoud9 order if the prices were to cross (for example, if one order is pegged to the NBBO Midpoint and the other is a Limit Order). If the CIoud9 Limit prices of two working orders cross, the system preferably attempts to pull back all orders routed to third party destinations, to maximize the size that can be traded internally. Residues from this match event are subsequently reassigned to external destinations. In yet another alternate embodiment, any trader may check a box to specify that they are willing to receive partial fills from the residue on a working order, and may optionally specify a minimum quantity that such fills must satisfy. In another alternate embodiment the orders entered within the system are placed in the execution engine in a suspended status while the liable block order is placed at a destination where a fill might be CTqiected. For example, the block can be placed on POSIT® for a scheduled cross while it remains in a suspended status on Cloud9. The system preferably maintains the effects of the suspended order on Cbud9 for purposes of the active symbol flags. Upon entry ofacontraorder, the system of this alternate embodiment preferably pulls back the liability by attempting to cancel the externalized order. When this cancel is successful (the order has not already been filled); the contras present flags are sent as in the preferred embodiment. While the embodiments shown and described herein are fully capable of achieving the objects of the subject mvention, it is evident that numerous alternatives, modifications, and variations will be apparent to those skilled in the art in light of the foregoing description. These alternatives, modifications, and variations are within the scope of the subject invention, and it is to be understood that the embodiments described herein are shown only for the purpose of illustration and not for the purpose of limitation. WE CLAIM ; 1. A system for communicating securities order information, the system comprising : network communication hardware 110 that electronically transmits and receives communication signals, wherein said communication signals comprise signals that represent: an order for a security, and an indication that a reasonably priced order for a security is present 215; an analytics module 160 in communication with said network communication hardware 110, wherein after said network communication hardware 110 receives a signal that represents a first order for a security from a first market participant, said analytics module 160 computes a block price range for said security, said block price range having an upper end and a lower end; a facilitator module 140 in communication with said network communication hardware 110 and said analytics module 160, wherein: said facilitator module 140 compares said price of said first order and said block price range and determines that said first order is reasonable if: said first order is a buy order and said price of said first order is equal to or greater than said lower end of said block price range, or said first order is a se]] order and said price of said first order is equal to or less than said upper end of said block price range; and if said facilitator module 140 determines that said price of said first order is reasonable, said facilitator module 140 instructs said network communication hardware 110 to transmit a signal to a second market participant which represents an indication that a reasonably priced order for said security is present 215, wherein said indication is silent on whether said reasonably priced order is a buy order or a sell order; and an order manager module 130 in communication with said network communication hardware 110, wherein after said network communication hardware 110 receives a signal that represents a second order for said security from said second market participant, said order manager module 130 determines whether said second order is contra to said first order and whether a price of said second order crosses said price of said first order, and if so, said order manager module 130 instructs a trade execution engine 50 to execute a trade comprising said first order and said second order. 2. The system as claimed in claim 1, wherein said analytics module 160 computes said block price range based on at least one of: National Best Bid, National Best Offer, recent market prices, and current market prices. 3. The system as claimed in claim 1, wherein said analytics module 160 computes said block price range based on recent volatility in said security. 4. The system as claimed in claim 1, wherein said analytics module 160 computes said block price range based on predicting a price range likely to occur within a first predetermined time period. 5. The system as claimed in claim 4, wherein said analytics module 160 recomposes said block price range at intervals of time approximately equal to said first predetermined time period. 6. The system as claimed in claim 1, wherein : Said communication signals comprise a signal that represents a notification to the first market participant that a nearly matching contra order is active within the system; and if said order manager module 130 determines that said second order is contra to said first order but said price of said second order does not cross said price of said first order, said facilitator module 140 determines if said second order is reasonable by comparing said price of said second order and said block price range to determine if said second order is a buy order and said price of said second order is equal to or greater than said lower end of said block price range, or said second order is a sell order and said price of said second order is equal to or less than said upper end of said block price range; and if said facilitator module 140 determines that said price of said second order is reasonable, said facilitator module 140 instructs said network communication hardware 110 to transmit a signal to said first market participant which represents a notification that a nearly matching contra order is active within the system. 7. The system as claimed in claim 6, wherein said facilitator module 140 instructs said network communication hardware 110 to transmit said signal to said first market participant representing said notification immediately after determining that said price of said second order is reasonable. 8. The system as claimed in claim 6, wherein: Said communication signals comprise a signal that represents a notification to the second market participant that a nearly matching contra order is active within the system; and if said order manager module 130 determines that said second order is contra to said first order but said price of said second order does not cross said price of said first order but said price of said second order is determined to be reasonable, said facilitator module 140 instructs said network communication hardware 110 to transmit a signal to said second market participant which represents a notification that a nearly matching contra order is active within the system. 9. The system as claimed in claim 8, wherein said facilitator module 140 instructs said network communication hardware 110 to transmit said signal to said second market participant representing said notification only after a second predetermined period of time has lapsed after said second order was received.

Documents:

3515-chenp-2005 abstract duplicate.pdf

3515-chenp-2005 abstract.pdf

3515-chenp-2005 assignment.pdf

3515-chenp-2005 claims duplicate.pdf

3515-chenp-2005 claims.pdf

3515-chenp-2005 correspondence (others).pdf

3515-chenp-2005 correspondence po.pdf

3515-chenp-2005 description (complete) duplicate.pdf

3515-chenp-2005 description (complete).pdf

3515-chenp-2005 drawings duplicate.pdf

3515-chenp-2005 drawings.pdf

3515-chenp-2005 form-1.pdf

3515-chenp-2005 form-18.pdf

3515-chenp-2005 form-26.pdf

3515-chenp-2005 form-3.pdf

3515-chenp-2005 form-5.pdf

3515-chenp-2005 others.pdf

3515-chenp-2005 pct search report.pdf

3515-chenp-2005 pct.pdf

3515-chenp-2005 petition.pdf