US20140304138A1 - System and Method for Matching One or More Incoming Order to a Standing Order Based on Multi-Level Allocation - Google Patents

Abstract

A method of order allocation is disclosed. The method includes receiving an incoming order, establishing an inner market representing a first portion of an order book which may be defined as a function of an inner market parameter, designating the first portion of the order book as a priority and allocating the first portion of the received incoming order based on the priority, establishing an outer market that represents a second portion of the order book that includes the remainder of the order book not represented by the inner market of the order book, assigning the received incoming order to one of the inner or outer markets as a function of the inner market parameter, allocating a first portion of the incoming order to the inner market utilizing a first-in, first-out (FIFO) algorithm, and allocating a second portion, in excess of the first portion, of the incoming order to the outer market using a pro-rata algorithm.

Description

CROSS REFERENCE TO RELATED APPLICATIONS
• This application is a continuation under 37 C.F.R. §1.53(b) and 35 U.S.C. §120 of U.S. patent application Ser. No. 12/246,796 filed Oct. 7, 2008, which is incorporated herein by reference in its entirety.
• This patent relates to co-pending U.S. patent application Ser. No. 11/729,578, titled “SYSTEM AND METHOD OF ALLOCATING AN INCOMING ORDER TO STANDING ORDERS”, filed on Mar. 29, 2007, the content of which is incorporated herein in its entirety for all purposes.
• This patent further relates to co-pending U.S. patent application Ser. No. 12/246,755, titled “SYSTEMS AND METHODS FOR MATCHING ONE OR MORE INCOMING ORDER TO A STANDING ORDER AS A FUNCTION OF AN INNER MARKET PARAMETER”, filed contemporaneously herewith; co-pending U.S. patent application Ser. No. 12/246,775, titled “SYSTEM AND METHOD FOR MATCHING ONE OR MORE INCOMING ORDER TO A STANDING ORDER BASED ON MULTIPLE ORDER PRIORITY”, filed contemporaneously herewith; co-pending U.S. patent application Ser. No. 12/246,818, titled “SYSTEM AND METHOD FOR MATCHING ONE OR MORE INCOMING ORDER TO A STANDING ORDER BASED ON MULTIPLE ORDER PRIORITY ALLOCATION”, filed contemporaneously herewith; and co-pending U.S. patent application Ser. No. 12/246,832, titled “SYSTEM AND METHOD FOR MATCHING ONE OR MORE INCOMING ORDER TO A STANDING ORDER BASED ON TIME ORDER PRIORITY ALLOCATION”, filed contemporaneously herewith, the contents of which are incorporated herein in their entirety for all purposes.
BACKGROUND
• An exchange provides one or more markets facilitating and/or enabling the purchase and sale of various types of products including financial instruments such as stocks, bonds, futures contracts, options, cash, and other similar instruments. Agricultural products and commodities are also examples of products traded on such exchanges. A futures contract is a product that is a contract for the future delivery of another financial instrument such as a quantity of grains, metals, oils, bonds, or cash. Generally, each exchange establishes, for each market provided thereby, a specification that defines at least the product(s) traded in the market, minimum quantities that must be traded, and minimum changes in price (e.g., tick size). For some types of products (e.g., futures or options), the specification further defines a quantity of the underlying product represented by one unit (or lot) of the product, and delivery and expiration dates.
• Products on an exchange are traded in an environment where the exchange provides a location for buyers and sellers to meet and negotiate a price for a quantity of a product. This environment may be an open outcry or on an electronic trading platform (e.g., an electronic exchange) where traders use software to send an order to the trading platform, or a combination of both. The order identifies the product, the quantity of the product the trader wishes to trade, a price at which the trader wishes to trade the product, and a direction of the order (i.e., whether the order is a bid to buy or an offer to sell). The trading platform may, in turn, allocate resting orders against: aggressor orders; stop orders; market orders; limit orders and/or any other type of known or foreseeable orders. Other products are traded in an over-the-counter market (e.g., where products are not listed on an exchange).
• In an electronic trading platform, a trading host on the trading platform monitors incoming orders and attempts to identify (i.e., match) one or more previously received orders stored in an order book database, wherein each identified order is contra to the incoming order and has a favorable price relative to the incoming order. In particular, if the incoming order is a bid then the identified order is an offer at a price that is identical to or less than the bid price. Similarly, if the incoming order is an offer at a particular price, the identified order is a bid at a price that is identical to or greater than the offer price. Unmatched orders are placed in an open order book to await receipt of a contra order, these waiting order are said to be “resting orders”.
• In one embodiment, the trading host calculates an open price or “indicative opening price” for a given product based on the overlap of orders (both bids and offers) present within the trading system prior to the market open. The mean price or indicative price will be updated periodically, i.e., every minute, every thirty seconds, every two minutes, etc. If the volume of the orders present within the trading system prior to the market open is such that the open price could be one of two prices, the price closest to the previous settlement price is selected to be the open price. If the trading host determines that overlap exists between bids and offers in a contract, the trading host totals the respective bids and the respective offers within the determined overlap range. The totaled bids are then analyzed by the trading host to determine where and at what price the highest total quantity on the bid “flips” or changes to a higher total quantity of the offer. The identified price is selected to be the indicative opening price. As previously discussed, this indicative opening price will be updated periodically, i.e., every minute, every thirty seconds, every two minutes, etc., until the opening of the market upon which the indicative opening price become the open price.
• Upon identification (matching) of a contra order, a minimum of the quantities associated with the identified order and the incoming order is matched and at least the minimum quantities of each of the identified and incoming orders become two halves of a matched trade that is sent to a clearinghouse. The trading host considers each identified order in this manner until either all of the identified orders have been considered or all of the quantity associated with the incoming order has been matched. If any quantity of the incoming order remains, an entry is created in the order book database and information regarding the incoming order is recorded therein.
• Traders access the markets on a trading platform using trading software that receives and displays at least a portion of the order book for a market, enables a trader to provide parameters for an order for the product traded in the market, and transmits the order to the trading platform. The trading software typically utilizes a graphical user interface to display at least a price and quantity of some of the entries in the order book associated with the market. The number of entries of the order book displayed is generally preconfigured by the trading software, limited by the trading platform, customized by the user or some combination thereof. Some graphical user interfaces display order books of multiple markets of one or more trading platforms. The trader may be an individual who trades on his/her behalf, a broker trading on behalf of another person or entity, a group, or an entity. Furthermore, the trader may be a system that automatically generates and submits orders.
• Regardless of the specific configuration of a particular trading platform, each of the trading platforms will typically include one or more matching algorithms, methods and schemes for allocating incoming orders to other incoming or resting orders. Because incoming orders will rarely be identical to, i.e., have the same quantities, prices and other variables, the other incoming or resting orders, the algorithms, methods and schemes are, for example, intended: to allocate orders to promote and/or reward market makers; to assure an equitable distribution of the incoming order; and to regulate the trading volume.
I. First-In First-Out (FIFO) Algorithm
• If the trading host identifies multiple orders having an identical price and which are contra to the incoming order, assuming that the price of the multiple orders is favorable to the price of the incoming order, the trading host allocates the quantity of the incoming order among such identified orders in accordance with prioritization and allocation algorithms as defined in the product specification. A time priority algorithm such as a first-in first-out (FIFO) allocation principle or algorithm considers each identified order sequentially in accordance with when the identified order was received. The quantity of the incoming order is matched to the quantity of the identified order received earliest, then quantities of the next earliest, and so on until the quantity of the incoming order is exhausted.
• FIG. 1 relates to an embodiment of a FIFO allocation algorithm that illustrates how an incoming order to sell one hundred (100) lots of a product at a price of 111.01/lot received by the trading host is allocated to standing orders. Table100 shows orders A to E that the trading host has identified in the order book that are contra to and that have a favorable price compared to the incoming order. Rows of the table100 show information regarding standing orders A through E in the order book. Acolumn104 shows the price of each order, acolumn106 shows the time each order was received, and acolumn108 shows the quantity requested by each order. Acolumn110 shows the portion of the quantity of the incoming order allocated to each of the standing orders A to E. In particular, the thirty (30) units requested in order A are allocated first because the order A has the most favorable price (111.03). The remaining seventy (70) units of the incoming order are allocated to the orders B to E in accordance with the time they were received because such orders are all at the same price. Therefore, the orders B and C are allocated fifty (50) and twenty (20) lots or units of the remaining seventy (70) lots or units of the incoming order, respectively. Because the quantity of the incoming order is exhausted, the remainder of the standing order C, and all of the standing orders D and E are not allocated any portion of the incoming order and remain a resting orders in the order book.
II. Pro-Rata Algorithm
• Some market specifications define the use of a pro-rata allocation algorithm or principle, wherein a quantity of an incoming order is allocated proportionally to each of plurality of identified orders.
• FIG. 2 relates to a pro-rata algorithm that illustrates an example of how an incoming order to sell one hundred (100) lots of a product with a price of 111.01/lot is allocated among standing orders. As with the example ofFIG. 1, the trading host has identified orders A to E as being contra (i.e., orders to buy) to the incoming order and having a favorable price. Further, thirty (30) lots of the incoming order are allocated to the thirty (30) lots requested in the order A first because this order has the highest bid price (i.e., is most favorable). Thereafter, the remaining seventy (70) lots are allocated to the orders B to E proportionally because these orders are at the same price. In particular, the trading host calculates (i) the total number of lots requested by the orders B to E (111 units) and (ii) calculates a proportion the quantity of each order comprises of the total.Column202 shows the proportion corresponding to each order. The trading host calculates the portion of the remaining quantity (70) of the incoming order to allocate to each of the orders B to E by multiplying the proportion of the total requested by the order and the quantity remaining. Typically, the trading host rounds the calculated quantity to an integer and whether the trading host rounds up, down, or to a nearest integer is determined by the specification for the market. In other embodiments, if the calculated quantity is a lot of one (1), then the trading host may be configured to not allocate any order during a first pass. Instead, these one (1) lot fills may be allocated, for example, on subsequent FIFO fills. Any quantity of the incoming order that remains after pro-rata allocation (e.g., because the trading host rounds down the calculated quantity) is allocated to any orders that have an unfilled quantity on a FIFO basis. For example, for the order B, the trading host multiplies the proportion of quantity requested by the order B (i.e., 45%) by the remaining quantity (70) and thus allocates thirty-two (32) lots of the remaining seventy (70) lots to this order.Column204 shows the quantity of the incoming order allocated to each of the orders A to E.
• Some trading platforms provide a priority to certain standing orders in particular markets. An example of such an order is the first order that improves a price (i.e., improves the market) for the product during a trading session. To be given priority, the trading platform may require that the quantity associated with the order is at least a minimum quantity. Further, some trading platforms cap the quantity of an incoming order that is allocated to a standing order on the basis of a priority for certain markets. In addition, some trading platforms may give a preference to orders submitted by a trader who is designated as a market maker for the product. Other trading platforms may use other criteria to determine whether orders submitted by a particular trader are given a preference. Typically, when the trading host allocates a quantity of an incoming order to a plurality of identified orders at the same price, the trading host allocates a quantity of the incoming order to any orders that have been given priority. The trading host thereafter allocates any remaining quantity of the incoming order to orders submitted by traders designated to have a preference, and then allocates any still remaining quantity of the incoming order using the FIFO or pro-rata algorithms.
III. Price Priority Algorithm
• FIG. 3 illustrates an example of allocating an incoming order to sell one hundred and ten (110) lots of a product where an order that improved the price is given priority. In this example, the order that is given priority must be for a quantity of more than twenty (20) lots and a maximum of fifty (50) lots are to be allocated to on the basis of priority. In addition, suppose that the order A improved the price and is for more than twenty (20) lots. For these reasons the order A is given priority and up to fifty (50) lots of any incoming order is allocated thereto before allocation to other orders at the same price. In particular, when an order for one hundred and ten (110) lots is received, fifty (50) lots are allocated to order A and the remaining quantity (60 lots) are allocated proportionally based on the quantity of standing orders that remains after the allocation to the order A based on the priority thereof as described above.
IV. Pro-Rata/Minimum Allocation Quantity Algorithm
• Pro-rata algorithms and principles utilized in some markets may require that an allocation provided to a particular order in accordance with the pro-rata algorithm must meet at least a minimum allocation quantity. Any orders that do not meet or exceed the minimum allocation quantity are allocated to on a FIFO basis after the pro-rata allocation (if any quantity of the incoming order remains). In the example illustrated inFIG. 4, acolumn402 shows the results of a pro-rata allocation of an incoming order of one hundred and ten (110) lots among orders A to E. Acolumn404 shows the results of applying a minimum allocation quantity that has a value of two (2) to the pro-rata allocation. In this example, the trading host rounds fractional quantities of the pro-rata allocation to the nearest integer. Because orders A and B would be allocated quantities that are less than the value of the minimum allocation quantity, the allocation to each of the orders A and B after applying the minimum allocation quantity is zero (0). The two (2) lots of the incoming order that were not allocated to orders A and B because of applying the minimum allocation quantity are allocated on a FIFO basis and because order A was received earliest, order A is allotted the entire remainder of two (2) lots (as shown in a column406). In an alternate embodiment, the two (2) lots of the incoming order that were not allocated to orders A and B, i.e., the residual, may be allotted to largest resting order that does not meet the minimum allocation quantity. In yet another alternate embodiment, all of the residual lots which, in this example are the two (2) lots of the incoming order that were not allocated to orders A and B, may be allocated to all of the remaining lots to the resting orders that did not receive an initial allocation.
• The pro-rata allocation algorithm may further include priority order functionality. Priority orders are orders that better a side of the market (a.k.a. market turner). Only one order per side of the market (buy side or sell side) can be a Priority order priority, although it is possible for neither side of the market to have a Priority order associated therewith. If a Priority order is present and the price elected or associated with the Priority order is selected, then the initial allocation is made to the Priority order. Once the quantity associated with the Priority order has been filled, the pro-rata allocation algorithm distributes the remaining orders as discussed in connection withFIG. 4. It will be understood that Priority orders may be associated with a minimum order quantity or volume and/or a maximum order quantity or cap.
V. Lead Market Marker (LMM) Allocation Algorithm
• In other embodiments a Lead Market Maker (LMM) allocation algorithm is disclosed. The LMM designation is given to individuals chosen to make a two-sided market in an assigned contract. The LMM designation conveys a guaranteed allocation percentage of incoming orders to the identified individuals as well as pricing concessions in return for satisfying market obligations. The LMM allocation algorithm has two versions: with and without Priority order functionality.
• VI. LMM Allocation Algorithm with Priority Order Functionality
• As previously discussed, Priority orders are orders that better a side of the market (a.k.a. market turner). Only one order per side of the market (buy side or sell side) can be a Priority order priority, although it is possible for neither side of the market to have a Priority order associated therewith. If a Priority order is present and the price elected or associated with the Priority order is selected, the Priority order is matched first regardless of whether or not the order belongs to a designated LMM. If the Priority order belongs to an LMM, the order will be matched first and will not be included in the following calculations: (1) if the LMM has a single order at an elected price level, it will match an LMM's allocation percentage N % of the remaining incoming order quantity. However, the matched quantity will not exceed the LMM's order quantity. (2) If the LMM has multiple orders at an elected price level, then the LMM quantity is aggregated and will match N % of the remaining incoming order quantity. Multiple LMM orders are matched on time priority basis until N % quantity allocation is fulfilled. All remaining resting orders at an elected price level (LMM as well) are matched according to time priority. In an embodiment, a user can lose their order priority (for Priority orders) when the order is changed under any one of the following scenarios: (i) an increase in quantity; (ii) a change in price; and (iii) a change in account number.
• VII. LMM Allocation Algorithm without Priority Order Functionality
• In another embodiment, if the LMM has a single order at an elected or associated price level, the LMM will match an LMM's allocation percentage N % of the incoming order quantity. However, the matched quantity will not exceed the LMM's order quantity. Alternatively, if the LMM has multiple orders at an elected price level, then the LMM quantity is aggregated and will match N % of the incoming order quantity. Multiple LMM orders are matched on Time Priority basis until N % quantity allocation is fulfilled. However, the LMM's allocated match quantity cannot exceed their aggregated order size quantity. All remaining resting orders at an elected price level (LMM included) are matched according to Time Priority. LMM allocation will not apply until after the market opening sequence is complete, with opening matching occurring on a FIFO basis. The opening price is calculated by maximizing match quantities in relation to buy/sell pressure.
• In other embodiments, an order matching algorithm may include a distributed allocation strategy that utilizes time priority elements for market-making orders and/or orders within a tick of the market (provided no resting orders remain at the same price). In operation, orders are initially processed on a time priority basis, the remaining resting orders are then allocated according to the following process: (1) each resting order at a given price receives a one lot allocation unless there is insufficient quantity and then the one lot allocation is conducted randomly; (2) the remaining resting orders are filled based on quantity allocation; and (3) any remaining quantity are allocated to the largest outstanding order or orders.
• U.S. Pat. No. 7,225,153 discloses digital options having demand-based, adjustable returns, and trading exchange therefor. In particular, the patent discloses methods and systems for trading and investing in groups of demand-based adjustable return (“DBAR”) contingent claims, including digital options, and for establishing markets and exchanges for such claims. The advantages of the discloses methods and systems, as applied to the establishment and operation of a DBAR digital options exchange, include the ability to offer investments whose profit and loss scenarios are comparable to those for digital options or other derivatives in traditional securities markets, without the need for options or derivatives sellers or order-matching of conventional markets. A DBAR digital options exchange of the disclosed methods and systems can also offer conditional investments, or limit orders, in which an investment in a state of a DBAR contingent claim (such as the price of an underlying asset or index) can be executed or withdrawn in response to the implied probability of the occurrence of that state.
• U.S. Pat. No. 7,233,923 discloses systems and methods for matching desired purchases and sales of mis-matched items. The disclosed systems and methods compare the requirements of potential buyers and sellers of items against those of sets of other potential buyers and sellers of items to construct transaction sets that will enable an optimal transaction or set of transactions for the parties involved. The disclosed systems and methods may be used to match purchases and sales of any items, such as goods, services, financial instruments, and property interests (e.g., ownership interests in real property, possessory interests in personal property, etc.).
• Allocation algorithms used by the trading host for a particular market may affect the liquidity of the market. Specifically, some allocation algorithms may encourage traders to submit more orders, where each order is relatively small. Other allocation algorithms may encourage a trader to use an electronic trading system that can monitor market activity and submit orders on behalf of the trader very quickly and without intervention. As markets and technologies available to traders evolve, the allocation algorithms used by trading hosts must also evolve accordingly to enhance liquidity and price discovery in markets.
BRIEF DESCRIPTION OF THE FIGURES
• FIG. 1 illustrates an example of allocating an incoming order to standing orders using a known FIFO algorithm;
• FIG. 2 illustrates an example of allocating an incoming order to standing orders using a known pro-rata algorithm;
• FIG. 3 illustrates an example of allocating an incoming order to standing orders in accordance with a known priority algorithm;
• FIG. 4 illustrates an example of allocating an incoming order to standing orders in accordance with a known pro-rata algorithm that imposes a minimum allocation quantity;
• FIG. 5 illustrates an embodiment of how a trading host may process an incoming order;
• FIG. 6A illustrates an embodiment of how a quantity of an incoming order may be allocated to standing orders;
• FIG. 6B illustrates another embodiment of how a quantity of an incoming order may be allocated to standing orders;
• FIG. 6C illustrates yet another embodiment of how a quantity of an incoming order may be allocated to standing orders;
• FIG. 7 illustrates an embodiment of how a quantity of an incoming order may be allocated to standing orders that have priority;
• FIG. 8 illustrates another embodiment of how a quantity of an incoming order may be allocated to standing orders that have priority;
• FIG. 9 illustrates an embodiment of allocating a quantity of an incoming order to orders submitted by traders who are given a preference;
• FIG. 10 illustrates an embodiment of how a quantity of an incoming order may be allocated to standing orders using a FIFO algorithm;
• FIGS. 11A to 11C illustrate an embodiment of how a quantity of an incoming order may be allocated to standing orders using a pro-rata algorithm;
• FIG. 12A illustrates an example of multiple standing orders having priority;
• FIG. 12B illustrates another example of multiple standing orders having priority;
• FIGS. 13A and 13B illustrate an example of the priority of standing orders being changed upon receipt of an additional order;
• FIGS. 14A and 14B illustrate an example of the priority of standing orders being revoked;
• FIG. 15 illustrates an example of allocation of an incoming order to resting orders in accordance with priority, preferences, FIFO allocation, and pro-rata allocation;
• FIG. 16A illustrates an example of allocation of an incoming order according to an inner/outer market allocation algorithm;
• FIGS. 16B and 16C illustrate tabular examples of allocation of an incoming order according to an inner/outer market allocation algorithm;
• FIG. 17 illustrates an example of allocation of an incoming order according to a priority pro-rata with multiple order priority algorithm;
• FIG. 18 illustrates an example of allocation of an incoming order according to a multi-level priority algorithm;
• FIG. 19 illustrates an example of allocation of an incoming order according to a multiple order priority algorithm;
• FIG. 20 illustrates an example of allocation of an incoming order according to a timed order priority algorithm; and
• FIGS. 21A and 21B illustrates embodiments of systems, components and devices that may implement the disclosed allocation algorithms, methods and processes.
DETAILED DESCRIPTION
• The present disclosure relates to systems, methods, process and devices for matching and allocating orders in an electronic trading environment.
I. First-In First-Out (FIFO)/Pro-Rata Split
• In one embodiment a matching algorithm may be configured to allocate an incoming order according to a combination of FIFO and pro-rata principles. For example, a portion of an incoming order may be allocated based on FIFO principles while the remainder of the incoming order may be allocated based on pro-rata principles.
• FIG. 5 illustrates aFIFO matching algorithm500 detailing a process by which a trading host may allocate an incoming order to identified orders. Atblock502, the trading host may wait, or continue to wait, to receive an order that is either a bid or an offer to buy or sell, respectively, a quantity of a product.
• Atblock504, in response to receiving an order, the incoming order may be validated by theFIFO matching algorithm500. Typically, validation includes verifying that the trader submitting the received order is authorized to trade in the market in which the product is traded, and that the incoming order conforms to the specifications for the market in which the product trades.
• Atblock506, in response to a determined invalid order, theFIFO matching algorithm500 posts an error to the trading software utilized by the trader submitting the order. At this point, theFIFO matching algorithm500 and the trading host returns to theblock502 to wait for another incoming order.
• Atblock508, in response to a determined valid order, theFIFO matching algorithm500 checks or queries a database (which may be thedatabase2116 shown inFIG. 21A) maintained by the trading host to determine if an order book (i.e., a market) exists for the incoming order. If the market does not exist, theFIFO matching algorithm500 proceeds to block510, otherwise theFIFO matching algorithm500 proceeds to block512.
• Atblock510, theFIFO matching algorithm500 may create a new order book for the market. Atblock514, theFIFO matching algorithm500 may create an entry in the newly created order book for the incoming order. TheFIFO matching algorithm500 and/or the trading host thereafter returns to theblock502 to wait for another order to arrive.
• Atblock512, the specifications for the market such as, for example, traded product, delivery month, strike price, etc., may be obtained or determined. For example, theFIFO matching algorithm500 queries or reads information such as the specifications for the market from a database (seedatabase2116 shown inFIG. 21A) or file structure (seememory2120 shown inFIG. 21B) stored or maintained on computers or terminals utilized by the trading platform. In some embodiments, the database that stores the specifications for the market may also store the order book. In other embodiments, separate databases may be utilized to store the specifications for the market and the order book. In still other embodiments, the specifications for the market may be obtained by theFIFO matching algorithm500 via a network and/or Internet or other secure connection to a remote system.
• Atblock516, theFIFO matching algorithm500 sets a value Q_remainingto equal the quantity of products specified in the incoming order received atblock502.
• Atblock517, theFIFO matching algorithm500 queries the order book, which may have been established atblock510, to determine the existence of any orders, having a favorable price, contra to the incoming order. If no such orders are identified, theFIFO matching algorithm500 proceeds to theblock514 and creates an entry for the incoming order in the order book. If an existing order having a favorable price, contra to the incoming order is identified, then theFIFO matching algorithm500, atblock518 may set a FIFO percentage FIFO_percentthat represents the percentage of the incoming order that should be allocated using FIFO principles.
• TheFIFO matching algorithm500 may determine the value FIFO_percentby querying the database (seeFIG. 21A), obtaining the value from another communicatively connected system, receiving an input from a user or any other known or foreseeable means. In some markets, FIFO_percentmay be fixed and defined by the market specification. In other markets, FIFO_percentmay be varied during the course of the trading session in accordance with metrics associated with the market. In other embodiments, FIFO_percentis determined in accordance with a measure of activity in the market.
• Atblock518, theFIFO matching algorithm500 may utilize the criteria of the market specification to determine the value of FIFO_percent. For example, in some embodiments, theFIFO matching algorithm500 may set a first, large value of FIFO_percentwhen there are many orders in the market, and a second, low (or lower) value of FIFO_percentwhen there are few orders in the market. In an embodiment, a high value of FIFO_percentmay be utilized if the sum of the quantities associated with the orders in the market at the best price is high; otherwise, a relatively low value of FIFO_percentmay be utilized. In another embodiment, the FIFO_percentmay be determined dynamically as the number of orders in the market or the quantities associated with the orders varies. It will be understood, that other metrics and indicators of market activity may be utilized in other embodiment to determine the value of FIFO_percent.
• Atblock520, theFIFO matching algorithm500 may determine Q_FIFO, that is the quantity of the incoming order that is to be allocated using a FIFO algorithm, by multiplying Q_remainingwith FIFO_percent. It should be apparent that if FIFO_percentis one-hundred percent (100%) then theFIFO matching algorithm500 implemented by the trading host exclusively utilizes FIFO principles to allocate the quantity of the incoming order. In addition, if the value of FIFO_percentis zero percent (0%) then none of the quantity the incoming order is allocated utilizing the FIFO algorithm and the entire quantity is allocated using pro-rata principles and/or algorithms described above or other algorithms designated for the market.
• Atblock522, theFIFO matching algorithm500 identifies and selects all of the orders from the order book that have the best price and that have not had any quantity of the incoming order allocated thereto.
• Atblock524, theFIFO matching algorithm500 allocates Q_remainingthe quantity of the product specified by the incoming order. Upon completion of the allocation, Q_remainingis updated to reflect the allocated quantity.
• Atblock526, theFIFO matching algorithm500 updates the order book corresponding to each of the identified and selected orders discussed in connection withblock522 to reflect the allocated allocation received from the incoming order.
• Atblock528, theFIFO matching algorithm500 and the trading host reports, for each of the identified and selected orders to which a portion of the incoming order was allocated, a match between the selected order and the incoming order to the clearinghouse, the trader who submitted the selected order, and the trader who submitted the incoming order. In an embodiment, the match may be recorded in a database maintained by the trading platform.
• Atblock530, theFIFO matching algorithm500 may check to determine if the order quantity represented by Q_remainingis greater than zero. If the determined value of Q_remainingis greater than zero (i.e., the quantity of the incoming order has not been exhausted) then, theFIFO matching algorithm500 proceeds to theblock517 to determine the existence of any remaining orders, having a favorable price, contra to the incoming order. If the determined value of Q_remainingequals than zero (0), then theFIFO matching algorithm500 may proceed to block502 to wait for another incoming order.
• A. Price Allocation Algorithm
• In one embodiment a matching algorithm may be configured to allocate an incoming order according to a price priority criterion in conjunction with FIFO and pro-rata principles. For example, an incoming order or a portion of an incoming order may be identified as having a best or high price and may, in turn, be designated as a price priority order. Lots may then be initially allocated based on the price priority while the remainder of the incoming order may be allocated based on, for example, a preference criterion, FIFO and/or pro-rata principles.
• FIG. 6A illustrates aprice allocation algorithm600 that may be implemented at the block524 (seeFIG. 5) of theFIFO matching algorithm500. Theprice allocation algorithm600 details another matching or allocation scheme for allocating or distributing the orders represented by Q_remainingaccording to the best price.
• Atblock602, theprice allocation algorithm600 may determine whether more than one (1) order has been selected. If a single order is selected, then atblock604 theprice allocation algorithm600 determines a quantity of the incoming order that is to be matched Q_matchedto the contra or resting order. Q_matchedmay be calculated determining the minimum of the quantity of product specified by the selected order Q_orderand the quantity of the incoming order that remains to be allocated Q_remaining.
• Atblock606, theprice allocation algorithm600 updates the value Q_remainingby subtracting the value Q_matched.
• Atblock607, theprice allocation algorithm600 updates the quantity of the selected order that remains Q_orderby subtracting Q_matched. Upon completion of the processing represented by theblock607, it will be apparent one of the values Q_matchedor Q_remainingwill equal zero (0). Theprice allocation algorithm600 may proceed to block610 where the matching or allocation to the selected order is complete and other orders may be considered.
• Atblock611, theprice allocation algorithm600, allocates the quantity of the incoming order Q_remainingto the selected orders in accordance with any priority criteria defined by the market specification for the product.
• Atblock612, theprice allocation algorithm600, determines if any quantity of the incoming order represented by Q_remainingremains unfilled. If none remain, theprice allocation algorithm600 may proceed to block610 where the matching or allocation to the selected order is complete and other orders may be considered.
• Atblock613, theprice allocation algorithm600, if Q_remainingis greater than zero, allocates the remaining quantity to standing orders submitted by traders designated to have a preference.
• Atblock614, theprice allocation algorithm600 determines if any quantity of the incoming order represented by Q_remainingremains to be allocated. If none remain, theprice allocation algorithm600 may proceed to block610 where the matching or allocation to the selected order is complete and other orders may be considered.
• Atblock616, theprice allocation algorithm600 allocates the remaining quantity of the incoming order in a number of lots up to the value Q_FIFOof the incoming order to the selected orders using the FIFO algorithm. Upon completion of the allocation, theprice allocation algorithm600 adjusts the values Q_FIFOand Q_remainingaccordingly.
• Atblock618, theprice allocation algorithm600 determines if any quantity of the incoming order represented by Q_remaining. If none remain, theprice allocation algorithm600 may proceed to block610 where the matching or allocation to the selected order is complete and other orders may be considered.
• Atblock620, theprice allocation algorithm600 allocates the remaining quantity Q_remainingutilizing the pro-rata algorithm.
• At theblock610 theprice allocation algorithm600 returns to the implementing block or point, in thepresent example block524 shown inFIG. 5.
• B. Alternate Price Allocation Algorithm
• In one embodiment a matching algorithm may be configured to allocate an incoming order according to a preference criterion in conjunction with FIFO and pro-rata principles. For example, an incoming order or a portion of an incoming order may be identified as having a best or high price and may, in turn, be designated as a price priority order. Lots may then be initially allocated based on the price priority while the remainder of the incoming order may be allocated based on a preference criterion associated with, for example, a preferred trader, and the remainder of the incoming order may be allocated based on FIFO and/or pro-rata principles.
• FIG. 6B illustrates an alternateprice allocation algorithm630 which may be implemented in connection with theprice allocation algorithm600 illustrated inFIG. 6A. Specifically, alternateprice allocation algorithm630 allocates a remaining quantity of an incoming order to orders submitted by preferred traders before allocating any remaining quantity of the incoming order to orders based on priority. In this embodiment, blocks611 to613 of theprice allocation algorithm600 illustrated inFIG. 6A may be replaced byblocks632 to636 of the alternate price allocation algorithm,630. Thus, if theprice allocation algorithm600 illustrated inFIG. 6A determines that more than one order has been selected then processing proceeds to theblock632, which allocates the quantity Q_remainingof the incoming order to those selected orders submitted by preferred traders (e.g., according to the alternate price allocation algorithm630).
• Atblock634, the alternateprice allocation algorithm630 may determine if any quantity of the incoming order represented by Q_remaining. If none remain, the alternateprice allocation algorithm630 may proceed to block610 when the matching or allocation to the selected order is complete and other orders may be considered as illustrated inFIG. 6A.
• Atblock636, the alternateprice allocation algorithm630 allocates any quantity of the incoming order represented by Q_remainingon the basis of priority. The alternate price allocation algorithm then proceeds to block614 illustrated inFIG. 6A.
• C. Pro-Rata Allocation Algorithm
• In one embodiment a matching algorithm may be configured to allocate an incoming order according to a combination of pro-rata principles and FIFO principles. For example, an incoming order or a portion of an incoming order may be initially allocated based on pro-rata principles while the remainder of the incoming order may be allocated based on FIFO principles.
• FIG. 6C illustrates apro-rata allocation algorithm650 which may be implemented in connection with theprice allocation algorithm600 illustrates inFIG. 6A. In particular, the exemplarypro-rata allocation algorithm650 allocates a remaining quantity of the incoming order using the pro-rata algorithm before allocating a quantity of the incoming order using FIFO principles or algorithms. In this embodiment, theblocks652 to658 of thepro-rata allocation algorithm650 may be executed in place ofblocks616 to620 of theprice allocation algorithm600 illustratesFIG. 6A.
• Atblock652, thepro-rata allocation algorithm650 determines a value of Q_proratathat is to be allocated using the disclosed pro-rata algorithm or scheme. In an embodiment, Q_proratamay be the difference between Q_remainingand Q_FIFO. In another embodiment, Q_proratamay be developed in a manner similar to the development of the value Q_FIFOdescribed herein above.
• Atblock654, thepro-rata allocation algorithm650 allocates a quantity equal to Q_prorataof the incoming order to the selected orders using the pro-rata algorithm.
• Atblock656, thepro-rata allocation algorithm650 determines if any quantity of the incoming order remains and, if any quantity does remain, the algorithm may proceed to block658. If none remain, thepro-rata allocation algorithm650 may proceed to block610 when the matching or allocation to the selected order is complete and other orders may be considered as illustrated inFIG. 6A.
• Atblock658, thepro-rata allocation algorithm650 allocates the remaining quantity Q_remainingto the selected orders using the FIFO algorithm and may proceed to block610 illustrated inFIG. 6A.
• D. Priority Allocation Algorithm
• In one embodiment a matching algorithm may be configured to allocate an incoming order according to a price priority criterion in conjunction with FIFO and pro-rata principles. For example, an incoming order or a portion of an incoming order may be designated as a price priority order if it is determined to be one of the first three (3) orders that improve the price, and lots may be allocated a function of the price priority.
• FIG. 7 illustrates an embodiment of apriority allocation algorithm700 configured to allocate the remaining quantity Q_remainingof the incoming order to the selected contra or standing orders that have a priority associated therewith. Thepriority allocation algorithm700 may be implemented, for example, at theblock611 of theprice allocation algorithm600 illustrated inFIG. 6A. In this exemplary embodiment, thepriority allocation algorithm700 and the trading host may track a quantity of orders allocated based on a priority (N_priority) within a market during a trading session. For example, the market specification may define that the first three orders that improve the price during a trading session are to have priority. At the start of the trading session, the value of N_prioritymay be set to three (3) by thepriority allocation algorithm700 and trading host. N_prioritymay be decremented for each order that is allocated a quantity of an incoming order because of priority.
• Atblock704,priority allocation algorithm700 sets a value of a counter i to 1.
• Atblock706, thepriority allocation algorithm700 identifies any orders among the selected orders (e.g., those orders selected or identifiedblock522 illustrated inFIG. 5) that have a priority associated therewith.
• Atblock708, thepriority allocation algorithm700 sets a value to the number N of the priority orders identified atblock706.
• Atblock710, thepriority allocation algorithm700 determines if there are additional priority orders to be considered (i<=N) and if there is a remaining quantity of the incoming order to allocate (Q_remaining>0). If additional priority orders do not exist and/or no quantity of the incoming order remains to be allocated, thepriority allocation algorithm700 proceeds to block714 to continue allocating the remaining quantity of the incoming order.
• Atblock712, thepriority allocation algorithm700, upon determining the existence of additional orders and quantities to allocate, sets a value Q_orderto equal the quantity requested by the i^thpriority order. In an embodiment, the priority orders are selected sequentially in accordance with the time when each priority order was received. In another embodiment, the priority orders are selected in accordance with trader status. Other methods, scheme or algorithms of selecting the priority orders may be apparent to those with skill in the art.
• Atblock713, thepriority allocation algorithm700 determines if the value of Q_orderis greater than or equal to a minimum quantity min_prioritynecessary to have a portion of the incoming order allocated in accordance with the priority. The value of min_priorityis typically defined by the market specification. If the value of Q_orderis at least min_prioritythen processing proceeds to ablock716; otherwise processing proceeds to theblock717.
• Atblock716, upon determining that Q_orderis greater than or equal to the minimum quantity min_prioritythepriority allocation algorithm700 determines a value of Q_matchedto allocate to the i^thpriority order. Q_matchedis determined by calculating the minimum of the quantity of the i^thpriority order and the quantity that remains of the incoming order.
• Atblock718, Q_matchedis subtracted from Q_orderto calculate the quantity of the i^thpriority order that remains.
• Atblock720, thepriority allocation algorithm700 calculates the remaining quantity Q_remainingof the incoming order by subtracting Q_matchedfrom the Q_order.
• Atblock722, thepriority allocation algorithm700 sets the quantity remaining Q_remainingof the i^thpriority order to equal Q_order.
• Atblock717, the value of the counter i is incremented.
• E. Alternate Priority Allocation Algorithm
• In one embodiment a matching algorithm may be configured to allocate an incoming order according to a priority criterion based on a quantity value. For example, if the priority criterion is based on a quantity of five orders, lots may be allocated to each of the five orders according to FIFO, pro-rata or any other allocation principle or algorithm.
• FIG. 8 illustrates an alternatepriority allocation algorithm800 to control the allocation of an incoming order to one or more selected orders having an assigned priority such that a maximum of a predetermined quantity (Q_priority) of the remaining quantity Q_remainingare allocated to selected orders based on the priority.
• Atblock802, that alternatepriority allocation algorithm800 determines, queries a database or otherwise looks up the value of Q_prioritythat may be allocated in accordance with the priority. In an embodiment, the alternatepriority allocation algorithm800 and the trading host tracks the number of lots within a market and during a trading session that may be allocated based on a priority. The value of Q_prioritymay be stored in a database of the trading host. The value of Q_prioritymay be decremented as incoming orders are allocated to orders in accordance with a priority. In another embodiment, the alternatepriority allocation algorithm800 and/or the trading host sets Q_priorityto a predetermined value defined by the trading platform for a particular market.
• Atblock804, the alternatepriority allocation algorithm800 initializes the value of a counter i to 1.
• Atblock808, the alternatepriority allocation algorithm800 sets the value of a variable N to equal to the number of orders identified as having a priority associated therewith.
• Atblock810, the alternatepriority allocation algorithm800 determines: (1) the existence of any remaining priority orders (i<=N); (2) existence of any predetermined quantity of unallocated priority orders (Q_priority>0); and (3) existence of any remaining quantity Q_remainingthe incoming of order to be allocated.
• If any of the conditions (1) to (3) are false or otherwise do not occur and/or exist, the alternatepriority allocation algorithm800 proceeds to block814 and stores the quantity of Q_priority
• If all three conditions are true, the alternatepriority allocation algorithm800 proceeds to block812 and sets the value of Q_orderto the quantity of the i^thpriority order. In an embodiment, priority orders are considered in the order in which each priority order was received. In other words, the first priority order that may be selected by the alternatepriority allocation algorithm800 may be the priority order received earliest.
• Atblock813, the alternatepriority allocation algorithm800 compares the value of Q_orderwith the value min_priorityto determine if the value of Q_orderis sufficient to be considered for allocation. If the value of Q_orderis less than min_priority, then atblock817 the counter i increments.
• If the value of order is Q_orderat least min_prioritythen processing proceeds to ablock816; otherwise processing proceeds to ablock817.
• Atblock816, upon determination that the value of Q_orderat least min_priority, the alternatepriority allocation algorithm800 calculates a matched quantity Q_matchedthat is the amount of the remaining quantity Q_remainingto allocate to the i^thpriority order without exceeding Q_priority.
• Atblocks818,820, and822, the alternatepriority allocation algorithm800 subtracts the value Q_matchedfrom the values Q_order, Q_remainingand Q_priority, respectively to reflect the quantity allocated to the i^thorder.
• Atblock824, the alternatepriority allocation algorithm800 sets the quantity of the i^thorder to Q_order.
• Atblock817, the alternatepriority allocation algorithm800 may increment the counter i and proceeds to theblock810.
• Atblock814, the alternatepriority allocation algorithm800 stores the quantity of Q_priorityand allocation of the remaining quantity Q_remainingof the incoming order to the selected orders on the basis of priority is completed.
• F. Preferred Trader Allocation Algorithm
• In one embodiment a matching algorithm may be configured to identify orders received from a preferred trader as priority orders. For example, if an incoming order is recognized as an order from a trader identified as a preferred trader, then lots may be allocated to the priority orders according to FIFO, pro-rata or any other allocation principle or algorithm.
• FIG. 9 illustrates a preferredtrader allocation algorithm900 that allocates, on the basis of preferences provided to traders, the remaining quantity Q_remainingof the incoming order to the orders selected and identified at block522 (seeFIG. 5) and that have an unfilled quantity associated therewith.
• Atblock902, the preferredtrader allocation algorithm900 sets a counter value i to 1.
• Atblock904, the preferredtrader allocation algorithm900 identifies orders submitted by traders who are provided with, assigned or otherwise earn and/or acquire a preference and/or a priority status.
• Atblock906, the preferredtrader allocation algorithm900 sets the value of number of orders N to equal the identified orders atblock904.
• Atblock908, the preferredtrader allocation algorithm900 sets the value of the quantity variable Q to equal Q_remaining.
• Atblock910, the preferredtrader allocation algorithm900 analyzes the order book to determine if orders from one or more preferred traders remain and the quantity of the incoming order remains. If no further orders from preferred traders remain and/or none of the quantity of the incoming order remains then, atblock914, the preferredtrader allocation algorithm900 continues allocating any remaining quantity of the incoming order.
• Atblock912, the preferredtrader allocation algorithm900 sets a value of the variable Q_orderto equal the quantity of the i^thorder identified atblock904.
• Atblock916, the preferredtrader allocation algorithm900 identifies a value for a variable (R) that represents a predetermined portion of the incoming order to which the preferred trader is entitled and that may vary from trader to trader. In an embodiment, if the preferred trader has submitted multiple orders at the same price, the trading host and/or the preferredtrader allocation algorithm900 insures that the total allocation to the preferred trader does not exceed the predetermined portion by adjusting the value of the variable R accordingly.
• Atblock918, the preferredtrader allocation algorithm900 determines a value that represents the quantity of the incoming order to be matched to the ith order matched quantity Q_matchedby calculating the minimum of the quantity of the i^thorder and the product of the value of the variable R and the value of the variable Q.
• Atblocks920 and922, the value of the matched quantity Q_matchedmay be subtracted from the values Q_orderand Q_remaining, respectively.
• Atblock924, the preferredtrader allocation algorithm900 sets the value of the i^thorder to equal the value stored by the variable Q_orderand updates the order database to reflect the match.
• Atblock926, the preferredtrader allocation algorithm900 increments the value of the counter i and processing may return to the analysis step represented theblock910.
• G. Selected Order FIFO Allocation Algorithm
• In one embodiment a matching algorithm may be configured to allocate lots to selected orders according to FIFO principles. For example, the matching algorithm may identify a number of orders based on a predefined criterion. Lots may, in turn, be allocated to the identified orders according to FIFO principles up to, for example, a given quantity.
• FIG. 10 illustrates a selected orderFIFO allocation algorithm1000 that shows how the quantity Q_FIFOof the incoming order is allocated to selected orders using the FIFO algorithm.
• Atblock1002, theFIFO allocation algorithm1000 sorts the selected orders in accordance with the time each order was received.
• Atblock1004, a counter variable i may be set or assigned a value equal to one (1).
• Atblock1006, theFIFO allocation algorithm1000 may set a value of the number of orders (N) variable to equal the actual, selected number of orders.
• Atblock1008, theFIFO allocation algorithm1000 sets a value of the quantity of the product specified Q_ordervariable to equal the actual quantity of the product specified in i^thorder of the sorted and selected orders.
• Atblock1010, the quantity of the product specified calculates the quantity of the incoming order to allocate to the i^thorder using a FIFO algorithm. For example, theFIFO allocation algorithm1000 sets the matched quantity Q_matchedto equal the minimum of the quantity of the incoming order that remains to be allocated (Q_remaining), the quantity of orders that may be allocated using the FIFO algorithm (Q_FIFO), and the quantity of the ith order (Q_order).
• Atblocks1012,1014, and1016, theFIFO allocation algorithm1000 subtracts the value of matched quantity Q_matchedfrom the values of Q_order, Q_remaining, and Q_FIFOrespectively, to reflect the quantity of the incoming orders that is allocated to the ith order.
• Atblock1018, the quantity that remains of the ith order may be set to Q_order.
• Atblock1020, theFIFO allocation algorithm1000 may increment the value represented by the counter variable i by one (1) thereby allowing theFIFO allocation algorithm1000 at theblock1008 to select the next order received for allocation.
• Atblock1022, theFIFO allocation algorithm1000 may branch or continue to theblock1008 if any quantity of the incoming order remains that can be allocated using the FIFO algorithm (Q_FIFO>0) and if another selected order need to be considers (i<=N). If these conditions are not satisfied, theFIFO allocation algorithm1000 returns to the calling block (e.g., theblock616 shown inFIG. 6A).
• H. Selected Order Pro-Rata Allocation Algorithm
• In one embodiment a matching algorithm may be configured to allocate lots to selected orders according to pro-rata principles. For example, the matching algorithm may identify a number of orders based on a predefined criterion. Lots may, in turn, be allocated to the identified orders according to pro-rata principles up to, for example, a given quantity.
• FIGS. 11A to 11C illustrate an exemplary selected orderpro-rata allocation algorithm1100 configured to allocate the remaining quantity Q_remainingof an incoming order to one or more selected orders in accordance with the pro-rata algorithm. For example, the selected orderpro-rata allocation algorithm1100 may represent, compliment and/or replace the processing executed atblocks614,616 and/or620 shown inFIG. 6A.
• Atblock1102, a sum of the quantities of selected orders Q_summay be calculated based on an actual/determined sum of the quantities of the selected orders.
• Atblock1104, the number of selected orders N may be determined and assigned.
• At block1106 a counter variable i is set or assigned a value equal to one (1).
• Atblock1108, the selected orderpro-rata allocation algorithm1100 may determine the quantity of the i^thselected order.
• Atblock1110, the selected orderpro-rata allocation algorithm1100 may calculate a proportion that the quantity of the i^thselected order comprises of the total of the quantities of the selected order Q_portion.
• Atblock1112, the counter variable i may increment or increase the counter value by one (1).
• Ablock1114, the selected orderpro-rata allocation algorithm1100 may compare the counter variable i to the number of selected orders N and, if the value of counter variable i is less than the value of the selected orders N continues, branches or returns to theblock1108. However, if the value of the counter variable i is greater than the value of the selected orders N, the selected orderpro-rata allocation algorithm1100 may proceed to the block1116 (seeFIG. 11B) where the value of the counter variable i is reset to equal one (1).
• FIG. 11B, which continues on fromFIG. 11A, depicts that the selected orderpro-rata allocation algorithm1100, atblock1118, calculates a value for the pro-rata quantity Q_proratathat is the minimum of the total of the quantities of the selected orders Q_sumand the remaining quantity Q_remaining.
• Atblock1120, the selected orderpro-rata allocation algorithm1100 sets the value of the variable matched to equal zero (0).
• Atblock1122, the value of a matched quantity Q_matchedwhich equals the quantity of the incoming order to be allocated to the ith selected order may be calculated by multiplying the proportion of the i selected order (Q_portion[i]) and the value of the pro-rata quantity Q_prorata. In an embodiment, the product of Q_portion[i] and Q_proratamay be rounded down to the nearest integer by truncating any fraction portion thereof. In another embodiment, the product may be rounded up to the next higher integer if the fractional portion of the product is greater than 0.5 and down to the preceding lower integer otherwise. Other rounding schemes that may be used are known to those skilled in the art.
• Atblock1124, the value of the matched quantity Q_matchedmay be compared to a value of the minimum allocation quantity (MAQ) that is defined for the market. For example, the value of the MAQ may be defined in the specification for the market. If the matched quantity matched is Q_greaterthan the value of the MAQ then atblock1126, the selected orderpro-rata allocation algorithm1100 may set Q_orderto equal the quantity associated with the i^thorder. Alternatively, if the value Q_matchedis less than the value of the MAQ, then the selected orderpro-rata allocation algorithm1100 may proceed to block1136.
• Atblock1128, the selected orderpro-rata allocation algorithm1100 adds the value of Q_matchedto the value of the variable matched which was initialized atblock1120.
• Atblock1130, the value of the variable Q_matchedmay be subtracted from the value of the variable Q_order.
• Atblock1132, the selected orderpro-rata allocation algorithm1100 may set the quantity of the i^thorder to equal the value of the variable Q_order.
• Atblock1134, the counter variable i may increment or increase the counter value by one (1).
• Atblock1136, the selected orderpro-rata allocation algorithm1100 may determine or evaluate if the value of the variable matched is less than the value of the variable Q_prorata, and if the value of the counter variable i is less than or equal to the number of selected orders (N). If both comparisons are determined to be true, then the selected orderpro-rata allocation algorithm1100 proceeds to block1122 to continue allocating the incoming order to the remaining selected orders. If one or both of the comparisons false (or un-true), then the selected orderpro-rata allocation algorithm1100 may proceed to block1138 (seeFIG. 11C) to allocate any quantity of the incoming order that was not allocated.
• FIG. 11C, which continues on fromFIG. 11B, depicts that the selected orderpro-rata allocation algorithm1100, atblock1138 may calculate a value of the variable remainder that is the difference between the quantity that could have been allocated using the prorata algorithm (Q_prorata) and the quantity that actually was allocated process steps or procedures represented by theblocks1122 to1136 (represented by the variable matched). The value of the variable remainder represents the quantities not allocated because of rounding or because a value Q_matchedwas less than the value of the minimum allocation quantity.
• Atblock1140, the selected orderpro-rata allocation algorithm1100 may sort the selected orders. In an embodiment, the selected orders may be sorted in accordance with the time when each order was received. In another embodiment, the selected orders may be sorted in accordance with the portion of the sum of the quantities of all of the selected orders represented by the quantity of each selected order.
• Atblock1142, the counter variable i may be reset to a value of one (1).
• Atblock1144, the selected orderpro-rata allocation algorithm1100 may evaluate if any quantity of the variable remainder needs to be allocated (i.e., that the value remainder is greater than 0), and if all of the selected orders have been considered (i.e., that value of the counter i is less than or equal to the value N). If both evaluations are true, the selected orderpro-rata allocation algorithm1100 may proceed to block1146. However, if one or more are false (or un-true), then the selected orderpro-rata allocation algorithm1100 may proceed to block1148. Atblock1148, the selected orderpro-rata allocation algorithm1100 may calculate the remaining quantity Q_remainingof the incoming order after the allocation process discussed in connection with theblocks1102 to1156. For example, the value of the variable Q_remainingwhich represents the remaining quantity, may be decreased by the value of the variable matched and the value of the variable remainder. Thereafter, process may resume with the selected orderpro-rata allocation algorithm1100 returning to point at which the pro-rata allocation algorithm was initiated, e.g., theblock620.
• Atblock1146, the value of the variable Q_ordermay be set to equal to any remaining quantity of the ith order (as sorted by the block1140).
• Atblock1149, the value of the variable Q_matchedmay be calculated or determined based on the minimum of the value of the variable Q_orderand value of the variable remainder.
• Atblock1150, the value of the variable Q_matchedmay be subtracted from the value of the variable Q_orderand the result, in turn, may be set as the value of Q_order.
• Atblock1152, the selected orderpro-rata allocation algorithm1100 may set the value of the variable remainder to equal the result of subtracting the value Q_matchedfrom the current value of the variable remainder.
• Ablock1154 set the quantity associated with i^thorder to the value Q_order.
• Atblock1156, the counter variable i may increment or increase the counter value by one (1) and the selected orderpro-rata allocation algorithm1100 may proceed to block1144.
II. Multiple Order Priority
• A. Priority for Orders that Improve a Market
• In one embodiment a matching algorithm may be configured to designate a predefined number of orders as priority orders if the orders are determined to improve the market and/or satisfy a minimum order quantity.
• Some embodiments allow more than one order to have priority.FIG. 12A shows a table1200 of exemplary standing orders in a market. For this example, the first three (3) orders that improve the market and that request an order quantity of at least twenty (20) lots are given priority. In addition, the maximum quantity of an incoming order that is allocated to each order based on the priority is fifty (50).
• Column1202 indicates the time of receipt of each order.Columns1204 and1206 show the price and quantity requested by each order listed incolumn1202, respectively. As shown in thecolumn1204, each of the orders has an identical price and this example assumes this is the best price for the market.
• Column1208 shows a quantity of each order filled based on priority. Although the earliest received order is order A, no portion of order A may be filled based on priority because this order does not meet the minimum quantity criterion of twenty (20) lots. Orders B to D are given priority because these are the first three orders to improve the market and that meet the minimum quantity criterion. Furthermore, up to fifty (50) lots of the orders C and D may be allocated based on priority.
• B. Priority as a Function of Min/Max Quantity
• In one embodiment a matching algorithm may be configured to designate a predefined number of orders as priority orders if the orders are determined to improve the size and/or volume of the market.
• FIG. 12B shows a table1210 that depicts how an exemplary market may provide priority to standing orders that improve the market based on a predefined quantity. For example, an order that includes a quantity of one hundred and twenty (120) lots may be distributed or allocated to one or more standing bid orders (orders A to E) if the standing orders satisfy a minimum quantity requirement of twenty (20) lots, then they are prioritized and may receive up to a maximum of fifty (50) lots per order.
• Column1211 identifies when each order A to E was received.Column1212 shows the price associated with each of the orders A toE. Column1214 shows the quantity requested by each of the orders A toE. Column1216 shows the distribution and/or allocation of the 120 lots to the prioritized orders A to E. Specifically, order A does not receive any priority because order A does not meet the minimum requirement. Thirty (30) lots of order B may be allocated on the basis of priority. A maximum quantity of fifty (50), out of the ninety-five (95) lots requested by order C, may be allocated on the basis of priority because fifty (50) is the maximum. All of the thirty (30) lots requested by order D may be allocated on a priority basis. Of the fifty (50) lots requested by order E, only ten (10) lots may be allocated based on priority because the quantity available for priority is exhausted. Note, if the trading host receives a second or subsequent matching incoming (sell) order having a quantity of at least one-hundred (100) lots, the trading host may allocate forty-five (45) lots of the incoming order to order C, forty (40) lots to order E, and then the remaining fifteen (15) lots to order A.
• C. Retention of Priority for Orders that Improve a Market
• In one embodiment a matching algorithm may be configured to designate an order as a priority order if the order is determined to improve the market, and the priority is maintained even if another order is received and is designated a priority order.
• Some embodiments may allow orders that are designated as priority orders for improving the market, may retain the designated priority even if the market is further improved by subsequent orders.FIG. 13A shows a table1300 of representative standing orders A to D to sell a product in a market. In this example, as many as three (3) orders that improve the market may be designated as priority orders and each priority order may be allocated the maximum quantity of fifty (50) lots, if the identified orders are for at least the minimum quantity of twenty (20) lots.
• Column1302 shows the time of receipt of each order.Columns1304 and1306 show the price and quantities of each order, respectively. In this example, the price of each of the orders A to D is identical.Column1306 shows the quantity of each of the orders A to D that is eligible to be allocated based on priority. In particular, orders B to D are designated as priority orders while order A does not satisfy the minimum quantity requirement and is therefore not designated a priority order. Thus, orders B to D may be allocated up to the maximum quantity of fifty (50) lots assuming that the initial contra order(s) includes a sufficient quantity for this allocation.
• FIG. 13B relates to table1300 shown inFIG. 13A. In particular,FIG. 13B illustrates a table1310 that expands on the table1300 and includes an order E that improves the market price. Order E may be designated as a priority order because it improves the market price and satisfies the minimum quantity requirements. Orders B and C retain the priority previously provided because the market in this example may have as many as three (3) standing orders that have priority. For the same reason, the priority provided to order D before the receipt of order E may be revoked. In some markets, a predetermined number of orders at the same price are given priority. In other markets, a predetermined number of orders at different price levels are given priority.
• Column1312 shows the quantities of the orders B, C, and E that may be filled based on priority. While the orders B and C may retain their designated priority, the trading host may first fill standing orders at a better price before filling the orders B and C even if such standing orders do not have priority.
• D. Aging of Order Priority
• In one embodiment a matching algorithm may be configured to designate an order as a priority order if the orders are determined to improve the market, and the priority is maintained and associated with the resting order for a predetermined period of time before expiring.
• In an embodiment, orders may be designated priority orders for a period of time or a specific duration or until the occurrence of a predefined event such as, for example, the close of the market. After expiration of the period of time, the designated priority may be revoked. In another embodiment, priority designations for all orders may be revoked after a period of time or duration. For example, the period of time may begin when an initial (first) order is designated as a priority order. The period of time may be fixed for the market (e.g., 30 seconds, 10 milliseconds) or may be dynamic and adjusted in accordance with market activity. Alternatively or in addition to, the period of time may vary by market and in some markets the period of time may be a sub-second interval.
• FIGS. 14A and 14B illustrate the aging of order priorities. Specifically,FIG. 14A shows a table1400 of standing orders A to D in a market at a time 10:00:04, wherein orders A to C, the first three (3) orders, were received at 10:00:01, 10:00:02, and 10:00:03, respectively. Orders A to C have been designated priority orders and allocated lots accordingly (as shown in column1402). The examples illustrated inFIGS. 14A and 14B require no minimum quantity to receive priority or cap the maximum quantity of the lots within an order that may be filled in accordance with the designated priority.FIG. 14B shows a table1404 that illustrates the status of the standing orders A to Data time 10:01:04.Column1406 illustrates an embodiment in which the priority designation associated with the orders A to C are revoked after a period of time, in this case after one (1) minute.
• E. Priority Including FIFO/Pro-Rata and Min/Max Quantity
• In one embodiment a matching algorithm may be configured to designate an order as a priority order if the order is determined to improve the market and/or satisfy a minimum order quantity. The priority order may have lots allocated to it according to FIFO principle up to a maximum quantity.
• FIG. 15 illustrates a table1500 that represent a plurality of standing orders A to G in an exemplary market. For example, it may be assumed that (1) the orders A to G are sorted by the time each order was received; (2) order A was received first, and (3) order A improved the market. Furthermore, it may be assumed that the market (1) utilizes a value of FIFO_percentof thirty percent (30%); (2) designates two (2) market improving orders as priority orders; (3) requires a minimum order quantity of twenty (20) lots; and (4) limits or caps the quantity of an order that may be allocated on the basis of the priority to fifty (50) lots. The market may further round down allocations based on the use of a pro-rata algorithm. The market illustrated by table1500 may provide a preference to the traders who submitted orders C and E. In particular, order C may be entitled to an allocation of up to ten-percent (10%) of a quantity of the incoming order that remains after priority allocation. Similarly, order E may be entitled to receive up to five-percent (5%) of the quantity of the incoming order that remains after allocation of the incoming order to any orders that have priority.
• For example, turning to the table1500 shown inFIG. 15, it may be assumed that the trading host receives an order to sell five hundred (500) lots at 111.01, and that the orders A to G are orders, within the same market, to buy five hundred and seventy-eight (578). It may be further assumed that each of orders A to G has an identical bid price that is better than or equal to 111.01.
• Column1502 indicates that orders A and B are designated priority orders and are allocated fifty (50) and thirty (30) lots, respectively, of the incoming order. Order A requests eighty (80) lots, however the market caps priority allocations to fifty (50).
• The quantity of the incoming order that remains after the priority allocation is four hundred and twenty (420) lots. Therefore, orders C and E are allocated forty-two (42) lots and twenty-one (21) lots, respectively, because of the provided ten-percent (10%) and five-percent (5%) allocations (see column1504).
• The quantity of the original incoming order remaining after the preference allocation is three hundred and fifty-seven (357) lots. Of the three hundred and fifty-seven (357) lots, the FIFO_percentof thirty percent (30%) or one-hundred and seven lots (107) may be allocated utilizing a FIFO algorithm. In this example, the FIFO_percentof lots to be allocated is rounded down. It will be understood that the FIFO_percentof lots to be allocated may be rounded up (to one-hundred and eight lots (108) in this example).
• Column1506 shows the allocation to each order A, C and D (order B was completely filed via the priority allocation).Column1508 shows that three hundred and twenty-eight (328) lots remain to be filled. The quantity of the incoming that remains to be allocated is two hundred and fifty (250) lots.Column1510 shows the allocation of three hundred and twenty-eight (328) lots represented by the quantity allocated to each order D toG. Column1512 shows the allocation of the remaining two hundred and fifty (250) lots in accordance with the pro-rata proportion. Because the pro-rata algorithm used by this market rounds down fractional allocation quantities, 2 lots remain to be allocated and these are allocated on a FIFO basis to order D as shown in acolumn1514.
• Although the embodiments and examples described above illustrate the use of a FIFO_percentvalue to determine the number of contracts that are allocated using the FIFO algorithm, it should be apparent that a ProRata_percentvalue may be used instead to determine the number of contracts that are allocated using the pro-rata algorithm. Furthermore, an incoming order may be allocated in accordance with the pro-rata algorithm first and then any quantity that remains thereafter may be allocated using the FIFO algorithm. In addition the incoming order may be allocated to standing contra orders using a combination of FIFO, pro-rata, multiple priority and multiple preferencing algorithms. Such algorithms may be combined with other allocation and order selection algorithms that may be apparent to those with skill in the art.
III. Inner/Outer Market Allocation
• In one embodiment a matching algorithm may be configured to identify an inner market and an outer market. Incoming orders within a market parameter, e.g., five (5) ticks, are identified as inner market orders and have lots allocated to them based on FIFO principles. All remaining orders are identified as outer market orders and have lots allocated to them based on pro-rata principles.
• FIG. 16A Illustrates an inner/outermarket allocation algorithm1600, the teaching and disclosure of which may be implemented at or in conjunction with the block524 (seeFIG. 5) of theFIFO matching algorithm500. The inner/outermarket allocation algorithm1600 details another matching or allocation scheme for segmenting orders and distributing or allocating contra orders represented by the value of Q_remainingaccording an inner market parameter IM_parameter. For example, in one embodiment, the inner/outermarket allocation algorithm1600 may include an IM_parameterset to equal five (5) ticks. In particular, the inner/outermarket allocation algorithm1600 identifies orders, within a predefined range e.g., within five (5) ticks between the contra bids and orders.
• The identified orders, in turn, may be designated as priority orders which represent an inner market. All other, undesignated orders may represent the outer market. The inner market parameter IM_parametermay be adjusted or defined based, for example, on market volatility. In a volatile market, IM_parametermay be a large value to account large changes in the market. Conversely, in a stable or more stable market, IM_parametermay be a smaller value because the changes may be less severe. Alternatively, IM_parametermay be adjusted throughout a period of time, for example, a trading session, in response to the volume of the market.
• In another embodiment, the inner market parameter IM_parametermay be the two (2) orders with the highest bid or value resting within the order book. In this example, the orders corresponding with the two highest bids may be designated as priority orders. As the highest bid changes over time, the two (2) orders allowed by IM_parametermay change and vary. Regardless of the value or variable represented by the inner market parameter IM_parameter, the inner/outermarket allocation algorithm1600 may be configured to allocate lots to inner market orders according to FIFO principles, while allocating lots to the remaining or outer market orders according to pro-rata principle.
• Returning toFIG. 16A, the disclosed inner/outermarket allocation algorithm1600 is assumed to be operable in connection with an existing market and order book such as, for example, the market discussed in connection withblocks502 to516 shown inFIG. 5. Atblock1602, the inner/outermarket allocation algorithm1600 may receive a new order which in this present embodiment may be an offer to sell a plurality of lots Q_order.
• Atblock1604, the quantity of lots remaining to be filled or allocated Q_remainingmay be set to include or equal the plurality of lots Q_orderplus any residual lots associated with Q_remaining. The quantity Q_remainingmay include any quantity of lots remaining from previous implementations of the inner/outermarket allocation algorithm1600.
• Atblock1606, the price associated with Q_ordermay be evaluated against the inner market parameter IM_parameterthat may be, for example, two (2) ticks between the bid and offer spread of contra orders. If the price associated with Q_orderis within the range defining the inner market, then the inner/outermarket allocation algorithm1600 may proceed along the path indicated by the arrow A. Alternatively, if the price associated with Q_orderis outside the range defining the inner market, then the inner/outermarket allocation algorithm1600 may proceed along the path indicated by the arrow B.
• Atblock1608, a predetermined number of incoming orders may be identified and designated priority orders based on the price associated therewith. In particular and for example, the first five orders that fall within two (2) ticks of the bid/order spread.
• Atblock1610, the quantity of lots remaining to be filled or allocated Q_remainingmay be allocated to the designated priority orders utilizing FIFO principles. The number of lots allocated to all of the designated priority orders, i.e. all of the inner market orders, may be identified as Q_FIFO.
• Atblock1612, the quantity of lots remaining to be filled or allocated Q_remainingmay be updated by subtracting the number of lots allocated utilizing FIFO principles Q_FIFOfrom the original quantity of lots remaining to be filled or allocated Q_remainingset atblock1604.
• Atblock1614, received orders having a price associated with Q_orderthat is outside the range defining the inner market less than or equal to the current high market price (as determined at block1606) and/or the quantity of lots remaining to be filled or allocated Q_remaining, may be allocated to the remaining orders within the order book according to pro-rata principles.
• FIGS. 16B and 16C show tables1640/1640′ and1680/1680′, respectively. In particular, the tables1640 and1680 illustrates multiple offers and bids corresponding to the orders A to H in a market organized and operable according to the inner/outermarket allocation algorithm1600 discussed above.
• FIG. 16B shows the tables1640 and1640′ which, in turn, depict the allocation of lots to contra orders A to H according to the inner/outermarket allocation algorithm1600. In particular,columns1642 and1644 of table1640 show time and prices associated with each of the orders A toH. Columns1646 and1648 show the contra bids and offers at a time interval one (1), the corresponding time intervals are indicated by the column heading “QTY @Time 1”. Similarly,columns1650 and1652 of table1640′ show the contra bids and offers at a time interval two (2), the corresponding time intervals are indicated by the column heading “QTY @Time 2”.
• The bids, attime interval 1, associated with orders D and E may be designated priority orders because at prices 111.05 and 111.04, respectively, they are within two (2) ticks of the nearest offer, i.e., they are within two (2) ticks of bid/offer spread. As new orders are received, the designated priority orders D and E may be filled or otherwise allocated lots according to FIFO principles. The remaining unallocated quantity Q_remainingfrom the example above, may be allocated according to pro-rata principles.
• The bids, attime interval 2, associated with orders F and G may designated priority orders because at prices 111.03 and 111.02, respectively, they are within two (2) ticks of the nearest offer, i.e., they are within two (2) ticks of bid/offer spread. As new orders are received, the designated priority orders F and G may be filled or otherwise allocated lots according to FIFO principles. The remaining unallocated quantity Q_remainingfrom the example above may be allocated according to pro-rata principles.
• It will be understood that many variations and alterations may be implemented to further refine the allocation of lots according to the inner/outermarket allocation algorithm1600. For example, the priority designation may or may not be a time-sensitive priority designation. If, for example, the priority designation associated with orders D and E were not a time-sensitive priority designation, these orders will retain priority over, and be allocated lots before, the priority designation associated with orders F and G. Thus, in operation, the priority orders D and E may be receive a priority allocation according to FIFO principles and orders F to H may receive an allocation according to pro-rata principles. The priority orders D and E may retain this priority until they have been filled or otherwise received their maximum allocation, and then the priority orders F and H may begin to receive their priority allocation until they have been filled or otherwise received their maximum allocation. In another variation, the FIFO principles utilized to allocate lots to priority orders may include, for example, FIFO_percentageto limit or otherwise govern the maximum volume or quantity allocated.
• If, for example, the priority designation associated with orders D and E were a time-sensitive priority designation, these orders will retain priority over, and be allocated lots before, for a given period and then may lose their priority designation to the orders F and G. At this point, and Q_remainingassociated with orders D and E may be, unless re-designated priority at a later time period, be allocated according to pro-rata principles. In yet another variation, a minimum order volume or limit Q_minmay be established at, for example, five (5) lots. In this example, the order D may no longer be eligible for FIFO allocation and may, instead, be allocated according to pro-rata principles.
• FIG. 16C shows the tables1680 and1680′ which, in turn, depict the allocation of lots to contra orders A to H according to the inner/outermarket allocation algorithm1600. In particular,columns1642 and1644 show time and prices associated with each of the orders A toH. Columns1682 and1684 of table1680 show the contra bids and offers at a time interval one (1), the corresponding time intervals are indicated by the column heading “QTY@Time 1”. . Similarly,columns1686 and1688 of table1680′ show the contra bids and offers at a time interval two (2), the corresponding time intervals are indicated by the column heading “QTY@Time 2”.
• The bids, attime interval 1, associated with orders D and F may designated priority orders because at prices 111.05 and 111.03, respectively, they represent the best or highest, i.e, the orders representing the greatest risk, standing bids within the order book. As new orders are received, the designated priority orders D and E may be filled or otherwise allocated lots according to FIFO principles. The remaining unallocated quantity Q_remainingfrom the example above, may be allocated according to pro-rata principles.
• The bids, attime interval 2, associated with orders E and F may be designated priority orders because at prices 111.04 and 111.03, respectively, they now represent the best or highest standing bids within the order book. As new orders are received, the designated priority orders E and F may be filled or otherwise allocated lots according to FIFO principles. The remaining unallocated quantity Q_remainingfrom the example above may be allocated according to pro-rata principles.
• IV. Priority Pro-Rata with Multiple Order Priority
• In one embodiment a matching algorithm may be configured to designate a predefined number of orders as priority orders and lots may be allocated to these priority orders according to pro-rata principles.
• FIG. 17 Illustrates a prioritypro-rata allocation algorithm1700, the teaching and disclosure of which may be implemented at or in conjunction with the block524 (seeFIG. 5) of theFIFO matching algorithm500. The prioritypro-rata allocation algorithm1700 details another matching or allocation scheme for allocating or distributing Q_remainingaccording to maximum volume cap. For example, in one embodiment, the prioritypro-rata allocation algorithm1700 may designate the first one-thousand (1000) lots or contracts as priority lots or contracts. The priority lots or contracts may, in turn, be allocated according to pro-rata principles. In an embodiment, if the quantity of an order that includes one or more of the priority contracts is reduced during the pro-rata allocation process, the order and/or the lots may retain priority. In another embodiment, the first one-thousand (1000) lots or contracts designated as priority lots or contracts may be determined based on the initial volume or initial number of lots or contracts, as opposed to a volume that could include subsequent new orders.
• Returning toFIG. 17, the disclosed prioritypro-rata allocation algorithm1700 is assumed to be operable in connection with an existing market and order book such as, for example, the market discussed in connection withblocks502 to516 shown inFIG. 5. Atblock1702, the prioritypro-rata allocation algorithm1700 may receive a new order which in this present embodiment may be an offer to sell a plurality of lots Q_order.
• Atblock1704, the total number of lots or contracts Q_totalpending within an order book for a given market may be updated. For example, the plurality of lots Q_ordermay be added to the current total number of lots or contracts Q_totalto determine a new value for Q_total. The value of Q_totalrepresents the total number of lots, regardless of the associated orders, that satisfy a given criterion such as price, time, etc.
• Atblock1706, the prioritypro-rata allocation algorithm1700 may evaluate the total number of lots or contracts Q_totalwith respect to a maximum volume or quantity cap or limit Q_cap. For example, Q_capmay be defined to equal one-thousand (1000) lots or contracts. Thus, if Q_totalis greater than Q_cap, that is, if total number of lots or contracts in an order book exceeds one-thousand, then the prioritypro-rata allocation algorithm1700 may proceed along the path indicated by the arrow A to theblock1716. If Q_totalis less than or equal to Q_cap, then the prioritypro-rata allocation algorithm1700 may proceed to theblock1708.
• Atblock1708, the prioritypro-rata allocation algorithm1700 may evaluate the total number of lots or contracts Q_totalwith respect to a minimum order volume or limit Q_min. If Q_orderis less than Q_min, that is, if number of lots or contracts in the received order books is less than a minimum amount, then the prioritypro-rata allocation algorithm1700 may proceed along the path indicated by the arrow B to theblock1716. If order is Q_greaterthan or equal to Q_min, then the prioritypro-rata allocation algorithm1700 may proceed to theblock1710.
• Atblock1710, the lots of contracts, under the one-thousand lot limit, in the received order Q_ordermay be identified or designated as priority lots or contracts.
• Atblock1712, priority orders, i.e., orders in include the first one-thousand lots or contracts may be filled first according to pro-rata principles Q_prorata. If any lots or contracts within one or more of the orders remain unfilled after the pro-rata allocation, these lots or contracts may retain priority until they are filled.
• Atblock1714, the quantity of lots remaining to be filled or allocated Q_remainingmay be updated by subtracting the number of lots allocated utilizing pro-rata principles Q_proratafrom the original quantity of lots remaining to be filled or allocated Q_remaining.
• Atblock1716, the quantity of lots remaining to be filled or allocated Q_remainingmay be allocated to the remaining orders within the order book according to FIFO, pro-rata or any other known allocation principles or variations thereof.
• Atblock1718, the prioritypro-rata allocation algorithm1700 may wait for new incoming orders and proceed along the path indicated by the arrow C to theblock1702 upon receipt of a new order.
V. Multi-Level Allocation
• In one embodiment a matching algorithm segment all of the resting orders and each segment may, in turn, be allocated lots according to a different principle. For example, one segment may be allocated lots according to FIFO principles, another segment according to pro-rata principles.
• FIG. 18 Illustrates amulti-level allocation algorithm1800, the teaching and disclosure of which may be implemented in connection with an existing market and order book such as, for example, the market discussed in connection withblocks502 to516 shown inFIG. 5. Themulti-level allocation algorithm1800 details another matching or allocation scheme for segmenting orders and distributing or allocating contra orders represented by the value of Q_remainingaccording one or more parameters or criteria.
• Atblock1802, themulti-level allocation algorithm1800 may receive a new order which in this present embodiment may be an offer to sell a plurality of lots Q_order.
• Atblock1804, the existing order book may be evaluated and organized, for example, any quantity of lots remaining to be filled or allocated Q_remainingmay be set to include or equal the plurality of lots Q_order. The quantity Q_remainingmay include any quantity of lots remaining from previous implementations of themulti-level allocation algorithm1800. The existing order book may be further evaluated and organized into multiple levels or segments. For example, in one embodiment, (I) the first five (5) orders at a new high market price may be assigned to a first (1^st) segment or level; (II) the remaining received orders or standing orders Q_remaininghaving a minimum order volume or limit Q_minabove a predetermined quantity may be assigned to a second (2^nd) segment or level; and (III) all other orders not allocated to first and second segments may be assigned to a third (3^rd) segment or level. It will be understood that orders may be segmented based on a variety of criteria, for example, orders may be assigned to second segment if they are orders having a price within a predefined number of ticks bid/order spread; if they are orders for a given quantity of lots; if they were received within a given time period, etc.
• Atblock1806, the first segment may be evaluated to determine if Q_remainingassociated therewith is greater than zero (0) indicating existence of orders within the first segment waiting to be filled, e.g., receive an allocation of lots. If Q_remainingis greater than zero, atblock1808, the orders within the first segment may be designated as priority orders and allocated lots according to the priority. The allocation may be accomplished according to any known allocation principle, algorithm or technique. Atblock1810, Q_remainingmay be updated to reflect the allocation of orders.
• Atblock1812, the second segment may be evaluated to determine if Q_remainingassociated therewith is greater than zero (0) indicating existence of orders within the first segment waiting to be filled, e.g., receive an allocation of lots. In particular, if Q_remainingremaining equaled zero (0) atblock1806 or another value atblock1810, then the value of Q_remainingmay be evaluated atblock1812. If Q_remainingis greater than zero (0), atblock1814, the orders within the first segment may be designated as priority orders and allocated lots according to the pro-rata principles. Alternatively, the allocation may be accomplished according to any know allocation principle, algorithm or technique. Atblock1816, Q_remainingmay be updated to reflect the allocation of orders.
• Atblock1818, the third segment may be evaluated to determine if Q_remainingassociated therewith is greater than zero (0) indicating existence of orders within the first segment waiting to be filled, e.g., receive an allocation of lots. In particular, if Q_remainingremaining equaled zero (0) atblock1812 or another value atblock1816, then the value of Q_remainingmay be evaluated atblock1818. If Q_remainingis greater than zero (0), atblock1820, the orders within the first segment may be designated as priority orders and allocated lots according to the FIFO principles. Alternatively, the allocation may be accomplished according to any know allocation principle, algorithm or technique.
• Atblock1822, Q_remainingmay be updated to reflect the allocation of orders.
• Atblock1824, themulti-level allocation algorithm1800 may wait for new incoming orders and proceed to theblock1802 upon receipt of a new order.
VI. Multiple Order Priority Allocation
• In one embodiment a matching algorithm may be configured to designate a predefined number of orders as priority orders if the orders are determined to improve the market.
• FIG. 19 Illustrates a multipleorder priority algorithm1900, the teaching and disclosure of which may be implemented in connection with an existing market and order book such as, for example, the market discussed in connection withblocks502 to516 shown inFIG. 5. The multipleorder priority algorithm1900 details another matching or allocation scheme for segmenting orders and distributing or allocating contra orders represented by the value of Q_remainingaccording one or more parameters or criteria.
• Atblock1902, the multipleorder priority algorithm1900 may receive a new order which in this present embodiment may be an offer to sell a plurality of lots Q_order.
• Atblock1904, the existing order book may be evaluated and organized, for example, any the quantity of lots remaining to be filled or allocated Q_remainingmay be set to include or equal the plurality of lots Q_order. The quantity Q_remainingmay include any quantity of lots remaining from previous implementations of the multipleorder priority algorithm1900. In one embodiment, for example, the first five (5) orders at a new high market price may be identified and designated as priority orders. It will be understood that orders may be identified based on a variety of criteria, for example, orders may be identified based on the value of Q_order; a period of time in which the order was received, etc.
• Atblock1906, the orders within the order book may be evaluated to determine whether or not each order has been designated as a priority order. If, for example, Q_remainingincludes one or more of the designated priority orders, then atblock1908, the priority orders (the first five (5) orders at a new high market price in the present example, may be allocated lots according to the priority. The allocation may be accomplished according to any know allocation principle, algorithm or technique. Atblock1910, Q_remainingmay be updated to reflect the allocation of orders.
• Atblock1912, if Q_remainingdid not include one or more of the designated priority orders or includes another value provided by theblock1810, then the lots associated with Q_remainingmay be allocated in accordance with pro-rata principles.
• Atblock1914, Q_remainingmay be updated to reflect the allocation of orders.
• Atblock1916, the multipleorder priority algorithm1900 may wait for new incoming orders and proceed to theblock1902 upon receipt of a new order.
VII. Timed Order Priority Allocation
• In one embodiment a matching algorithm may be configured to designate a predefined number of orders as priority orders if the orders are received within a given period of time. Order that are not received within the given period of time are distributed according to pro-rata principles.
• FIG. 20 Illustrates a timedorder priority algorithm2000, the teaching and disclosure of which may be implemented in connection with an existing market and order book such as, for example, the market discussed in connection withblocks502 to516 shown inFIG. 5. The timedorder priority algorithm2000 details another matching or allocation scheme for segmenting orders and distributing or allocating contra orders represented by the value of Q_remainingaccording one or more timing parameters or criteria.
• Atblock2002, the timedorder priority algorithm2000 may receive a new order which in this present embodiment may be an offer to sell a plurality of lots Q_order.
• Atblock2004, the quantity of lots remaining to be filled or allocated Q_remainingmay be set to include or equal the plurality of lots Q_order. The quantity Q_remainingmay include any quantity of lots remaining from previous implementations of the timedorder priority algorithm2000.
• Atblock2006, the price associated with Q_ordermay be evaluated to determine if it represents the new low price for the market. If the price is determined to be the new low market price, then atblock2008, Q_ordermay be designated as a priority order and a may be activated. The timer may be set for any time or priority period, for example, thirty (30) seconds, one (1) minute, etc.
• Atblock2010, the timedorder priority algorithm2000 may allocate new lots from contra received orders to satisfy or fill Q_order. For example, during the time or priority period, the timedorder priority algorithm2000 may allocate lots based on FIFO principles.
• Atblock2012, Q_remainingmay be updated to reflect the allocation of orders.
• Atblock2014, the predefined time or priority period may be evaluated against the elapsed time (initiated at block2008) to determine if the priority period has ended. If the priority period has not ended, timedorder priority algorithm2000 may return to theblock2010.
• Atblock2016, if Q_orderwas not associated with a new low market price as determined atblock2006 or the priority period has ended as determined atblock2014, then any lots associated with Q_remainingmay be allocated in accordance with pro-rata principles.
• Atblock2018, Q_remainingmay be updated to reflect the allocation of orders.
• Atblock2020, the timedorder priority algorithm2000 may wait for new incoming orders and proceed to theblock2002 upon receipt of a new order.
• The trading methods, allocation algorithms and variants thereof generally relate to information processing and optimization. In particular, the disclosed trading methods and allocation algorithms embody methods and processes for allocating resources based on criteria such as, for example, price, quantity, time of receipt, etc.
• FIG. 21A illustrates one embodiment of asystem2100 configured to implement one or more of the disclosed trading methods, allocation algorithms, etc. Thesystem2100 may includemultiple terminals2102 to2108 directly and/or indirectly in communication with anorder management terminal2110. For example, theterminals2102 and2104 may communicate with theorder management terminal2110 via theInternet2112, a wide area network (WAN) and/or other communication network. Theterminals2106 and2108 may communicate with theorder management terminal2110 via for example acommunication network2114 such as an Ethernet network, a wireless fidelity (WiFi) network and/or other communication network. Theorder management terminal2110 may, in turn, be in communication with adatabase2116 or other memory or storage device or medium. Thedatabase2116 may be configured to store, in an accessible manner, the information, algorithms, parameters, etc. necessary to implement and monitor the trading methods, allocation algorithms disclosed herein. Thedatabase2116 may be a separate device or logical construct or may be a portion of theorder management terminal2110.
• FIG. 21B illustrates one example of a logical configuration that may be implemented in theorder management terminal2110. For example, theorder management terminal2110 may include acommunication module2118 and amemory2120 in communication with aprocessor2124 via acommunication bus2122. Thememory2120 may include RAM, ROM, flash memory, or any other type of known storage medium. Moreover, thememory2120 may include thedatabase2116 stored thereon. Thecommunication module2118 may be a wireless communication module or may be a wired communication module.
• Theprocessor2124 may be a general purpose processor configured to execute the disclosed trading methods and allocation algorithms. Alternatively, theprocessor2124 may represent one or more application specific processor ormodules2124ato2124c. For example, themodule2124amay be a FIFO allocation module or processor; themodule2124bmay be a pro-rata allocation module or processor; and themodule2124cmay be a tracking module or processor for processing and updating the order status associated with each method and/or algorithm.
• It should be understood that various changes and modifications to the presently preferred embodiments described herein will be apparent to those skilled in the art. Such changes and modifications can be made without departing from the spirit and scope of the present invention and without diminishing its intended advantages. It is therefore intended that such changes and modifications be covered by the appended claims.

Claims (20)

1. An apparatus comprising:

a memory configured to store an order book; and

a processor configured to allocate a first order segment to a first portion of the order book based on a first criterion using a first allocation algorithm and allocate a second order segment to a second portion of the order book based on a second criterion using a second allocation algorithm.

2. The apparatus ofclaim 1, wherein the first criterion is different than the second criterion and based on priority, quantity, or time of receipt.

3. The apparatus ofclaim 1, wherein the processor is configured to allocate a third order segment to a third portion of the order book using a third allocation algorithm.

4. The apparatus ofclaim 3, wherein the third allocation algorithm is a first-in, first-out (FIFO) allocation algorithm defined by a FIFO percentage to determine the percentage of an incoming order to be allocated according to the FIFO algorithm.

5. The apparatus ofclaim 1, further comprising:

a communication device configured to receive data indicative of a order including the first order segment and the second order segment from one or more terminals over a communication network.

6. The apparatus ofclaim 1, wherein the communication network is wireless.

7. The apparatus ofclaim 1, wherein the first criterion is based on market activity, and the processor is configured to calculate a size of the first portion of the order book based on market activity.

8. The apparatus ofclaim 1, wherein the first allocation algorithm is a priority allocation algorithm.

9. The apparatus ofclaim 1, wherein the second allocation algorithm is a pro-rata allocation algorithm.

10. The apparatus ofclaim 1, wherein the pro-rata allocation algorithm includes a pro-rata percentage to determine a percentage of an incoming order to be allocated according to the pro-rata allocation algorithm.

11. A method comprising:

receiving an incoming order;

allocating a first order segment of the incoming order to a first portion of the order book based on a first criterion using a first allocation algorithm; and

allocating a second order segment of the incoming order to a second portion of the order book based on a second criterion using a second allocation algorithm.

12. The method ofclaim 1, wherein the first criterion is different than the second criterion and based on priority, quantity, or time of receipt.

13. The method ofclaim 1, further comprising:

allocating a third order segment to a third portion of the order book using a third allocation algorithm.

14. The method ofclaim 13, wherein the third allocation algorithm is a first-in, first-out (FIFO) allocation algorithm defined by a FIFO percentage to determine the percentage of the incoming order to be allocated according to the FIFO algorithm.

15. The method ofclaim 11, wherein the first criterion is based on market activity, and a size of the first portion of the order book is calculated based on market activity.

16. The method ofclaim 11, wherein the first allocation algorithm is a priority allocation algorithm.

17. The apparatus ofclaim 11, wherein the second allocation algorithm is a pro-rata allocation algorithm defined by a pro-rata percentage to determine a percentage of the incoming order to be allocated according to the pro-rata allocation algorithm.

18. A non-transitory computer readable medium including instructions configured to cause at least a processor to perform:

establishing a first order segment as a first portion of an order book;

establishing a second order segment as a second portion of the order book that is different than the first portion;

establishing a third order segment, wherein the third order segment is a third portion of the order book;

designating the first order segment as a priority portion;

receiving an incoming order;

assigning the incoming order to one of the first, second or third segments;

allocating a quantity of a second incoming order to the priority portion utilizing a priority allocation algorithm;

allocating a first remaining quantity of the second incoming order portion utilizing a pro-rata allocation algorithm; and

allocating a second remaining quantity of the second incoming order portion utilizing a first-in, first-out (FIFO) allocation algorithm.

19. The non-transitory computer readable medium ofclaim 18, wherein the pro-rata allocation algorithm includes a pro-rata percentage to determine a percentage of the incoming order to be allocated according to the pro-rata allocation algorithm.

20. The non-transitory computer readable medium ofclaim 18, wherein the FIFO allocation algorithm includes a FIFO percentage to determine a percentage of the incoming order to be allocated according to the FIFO algorithm.

Images