Transaction routing determination method and device

Abstract

The present invention provides a transaction routing determination method and device for use in the financial field. The method comprises: constructing a transaction center network, the transaction center network comprising a decision center node and multiple execution nodes, wherein the decision center node is connected to each execution node via edges, wherein the edges are paths; calculating the average network delay of different execution nodes based on their heartbeat data; fitting the BBO change frequency of different execution nodes based on their historical BBO data; fitting the current transaction costs of different execution nodes based on their historical transaction costs; calculating the transaction network weight of the execution node; and determining the path corresponding to the execution node corresponding to the minimum transaction network weight as the shortest path. The present invention can determine the transaction routing of the shortest path, improve the transaction execution success rate of the algorithmic strategy, and thus reduce transaction costs.

Description

Technical Field

[0001] It should be noted that the transaction route determination method and device of the present invention can be used in the financial field, and can also be used in any field other than the financial field. The application field of the transaction route determination method and device of the present invention is not limited. Background Art

[0002] For algorithmic trading in multiple markets, trading decisions rely on market data input from physically distributed exchanges. Therefore, a centralized network exists between a decision node (algorithmic strategy) and multiple execution nodes (exchanges). The path between the decision node and the execution node in this centralized network is the transaction route. However, there is currently a lack of an accurate solution for determining transaction routes. Summary of the Invention

[0003] An embodiment of the present invention provides a transaction routing determination method for determining the shortest path transaction routing, thereby improving the transaction execution success rate of the algorithmic strategy and reducing transaction costs. The method includes:

[0004] Constructing a transaction center network, wherein the transaction center network includes a decision center node and multiple execution nodes, and the decision center node is connected to each execution node through an edge, wherein the edge is a path;

[0005] Calculate the average network delay of different execution nodes based on the heartbeat data of different execution nodes;

[0006] Based on the historical BBO data of different execution nodes, the BBO change frequency of different execution nodes is obtained by fitting;

[0007] Based on the historical transaction costs of different execution nodes, the current transaction costs of different execution nodes are obtained by fitting;

[0008] For each execution node in the trading center network, the transaction network weight of the execution node is calculated based on the average network delay, BBO change frequency, and current transaction cost of the execution node;

[0009] The path corresponding to the execution node corresponding to the minimum transaction network weight is determined to be the shortest path.

[0010] An embodiment of the present invention provides a transaction route determination device for determining the shortest path transaction route, thereby improving the transaction execution success rate of the algorithmic strategy and reducing transaction costs. The device includes:

[0011] A transaction center network construction module is used to construct a transaction center network, wherein the transaction center network includes a decision center node and multiple execution nodes, and the decision center node is connected to each execution node via an edge, wherein the edge is a path;

[0012] The network delay mean calculation module is used to calculate the network delay mean of different execution nodes based on the heartbeat data of different execution nodes;

[0013] The BBO change frequency fitting module is used to fit the BBO change frequency of different execution nodes based on the historical BBO data of different execution nodes;

[0014] The current transaction cost fitting module is used to fit the current transaction costs of different execution nodes based on the historical transaction costs of different execution nodes;

[0015] The transaction network weight calculation module is used to calculate the transaction network weight of each execution node in the transaction center network based on the average network delay, BBO change frequency, and current transaction cost of the execution node;

[0016] The shortest path determination module is used to determine the path corresponding to the execution node corresponding to the minimum transaction network weight as the shortest path.

[0017] An embodiment of the present invention further provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned transaction route determination method when executing the computer program.

[0018] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the method for determining a transaction route is implemented.

[0019] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned transaction route determination method is implemented.

[0020] In an embodiment of the present invention, a transaction center network is constructed, comprising a decision center node and multiple execution nodes, wherein the decision center node is connected to each execution node via edges, where the edges are paths. The average network delay of each execution node is calculated based on its heartbeat data. The BBO change frequency of each execution node is fitted based on its historical BBO data. The current transaction cost of each execution node is fitted based on its historical transaction cost. For each execution node in the transaction center network, the transaction network weight of the execution node is calculated based on its average network delay, BBO change frequency, and current transaction cost. The path corresponding to the execution node with the minimum transaction network weight is determined as the shortest path. Compared to existing technologies, this method uses the average network delay, BBO change frequency, and current transaction cost to determine the transaction network weight of each execution node, thereby determining the shortest path and achieving the most appropriate transaction routing. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0022] Figure 1 Flowchart of a transaction routing determination method according to an embodiment of the present invention;

[0023] Figure 2 This is a flow chart for calculating the average network delay of an execution node in an embodiment of the present invention;

[0024] Figure 3 This is a flow chart of calculating the transaction network weight of an execution node in an embodiment of the present invention;

[0025] Figure 4 Another flow chart of calculating transaction network weight in an embodiment of the present invention;

[0026] Figure 5 Schematic diagram of a transaction route determination device according to an embodiment of the present invention;

[0027] Figure 6 Schematic diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] To make the purpose, technical solutions and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0029] In the description of this specification, the terms "include", "including", "have", "contain", etc. are all open terms, which mean including but not limited to. The descriptions with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. The order of steps involved in each embodiment is used to schematically illustrate the implementation of the present application, and the order of steps therein is not limited and can be appropriately adjusted as needed.

[0030] First, the terms used in the present invention are explained.

[0031] Trade routing: The same financial asset is traded on multiple exchanges in the financial markets. Financial institutions, driven by various trading objectives such as arbitrage, hedging, and client hedging, need to find the best price and volume across multiple exchanges' order books. The market depth and liquidity of exchange order books form the basis for algorithmic trading decisions and executions. The process by which trading strategies initiate trades and select trading venues is called trade routing.

[0032] Algorithmic trading, also known as automated trading or black-box trading, involves the use of electronic platforms to input trading instructions involving algorithms to execute pre-defined trading strategies. The algorithms incorporate numerous variables, including time, price, and volume, or, in many cases, orders are initiated by "robots" without human intervention. Algorithmic trading is widely used by investment banks, pension funds, mutual funds, and other buy-side institutional investors to manage market risk and shocks by breaking large trades into many smaller ones. Sell-side traders, such as market makers and some hedge funds, provide liquidity to the market, automatically generating and executing orders.

[0033] The inventors believe that whether transactions sent to exchanges can be executed at the expected price and volume is directly related to the paths between the decision nodes and the exchanges. Therefore, the present invention proposes a transaction routing solution.

[0034] Figure 1 FIG. 1 is a flow chart of a transaction routing determination method according to an embodiment of the present invention. Figure 1 As shown, including:

[0035] Step 101: construct a transaction center network, wherein the transaction center network includes a decision center node and multiple execution nodes, and the decision center node is connected to each execution node via an edge, wherein the edge is a path;

[0036] Step 102, calculating the average network delay of different execution nodes based on the heartbeat data of different execution nodes;

[0037] Step 103: Based on the historical BBO data of different execution nodes, obtain the BBO change frequency of different execution nodes by fitting;

[0038] Step 104: Based on the historical transaction costs of different execution nodes, obtain the current transaction costs of different execution nodes by fitting;

[0039] Step 105: For each execution node in the transaction center network, calculate the transaction network weight of the execution node through the average network delay, BBO change frequency, and current transaction cost of the execution node;

[0040] Step 106: Determine that the path corresponding to the execution node corresponding to the minimum transaction network weight is the shortest path.

[0041] Among them, the BBO (Best Bid and Offer) frequency of change: A bid or ask / offer order represents a trader's willingness to buy or sell, including the price (bid or ask / offer) and quantity they are willing to accept. The highest bid and lowest ask prices are called the best bid (ask) price / quote. Market quotation includes both the best bid and ask prices, also known as the BBO (Best Bid and Offer).

[0042] Transaction cost refers to the handling fee charged by the exchange. The higher the fee, the greater the weight value.

[0043] In step 101, a trading center network is constructed. The trading center network includes a decision center node and multiple execution nodes. The decision center node is connected to each execution node via edges, which are paths. The execution nodes can be exchanges. The trading center network is obtained by connecting trading strategies to exchanges for network abstraction. The shortest path is calculated based on the average network delay, the frequency of BBO changes, and the current transaction cost. In the case of complex multi-path orders, the optimal path is selected across multiple exchanges.

[0044] In step 102, the network delay averages of different execution nodes are calculated based on the heartbeat data of different execution nodes; Figure 2 The flowchart of calculating the average network delay of execution nodes in an embodiment of the present invention is shown in FIG. 1 . In one embodiment, the average network delay of different execution nodes is calculated based on the heartbeat data of different execution nodes, including:

[0045] Step 201, processing the heartbeat data of different execution nodes by a time series method to obtain a heartbeat data sequence;

[0046] Step 202: Send the heartbeat data sequence to a stationary model to obtain a mean network delay. The stationary model is obtained by training historical heartbeat data and historical network delay mean values.

[0047] Among them, the stable model can be a fitting model or a deep neural network model, and there is no restriction here.

[0048] In step 103, based on the historical BBO data of different execution nodes, the BBO change frequency of different execution nodes is obtained by fitting; the fitting here can be a regression test fitting, of course, it can also be other fittings, which are not limited here.

[0049] In step 104, based on the historical transaction costs of different execution nodes, current transaction costs of different execution nodes are obtained by fitting; the fitting here can be a regression test fitting, of course, it can also be other fittings, which are not limited here.

[0050] In step 105, for each execution node in the transaction center network, the transaction network weight of the execution node is calculated based on the average network delay, BBO change frequency, and current transaction cost of the execution node; Figure 3 The flowchart of calculating the transaction network weight of an execution node in an embodiment of the present invention is as follows. In one embodiment, for each execution node in the transaction center network, the transaction network weight of the execution node is calculated based on the average network delay, BBO change frequency, and current transaction cost of the execution node, including:

[0051] Step 301: For each execution node, query the correspondence between the average network delay and the transaction cost, obtain the transaction cost corresponding to the average network delay of the execution node, and determine the first weight corresponding to the transaction cost corresponding to the average network delay; for example, when the transaction cost is in the range of 0 to a, the first weight is A; when the transaction cost is in the range of a to b, the first weight is B, and so on.

[0052] Step 302: For each execution node, query the correspondence between the BBO change frequency and the transaction cost, obtain the transaction cost corresponding to the BBO change frequency of the execution node, and determine the second weight corresponding to the transaction cost corresponding to the BBO change frequency;

[0053] Step 303: For each execution node, determine a third weight corresponding to the current transaction cost;

[0054] Step 304: Calculate the transaction network weight of the execution node based on the first weight, the second weight, and the third weight.

[0055] In practice, the BBO change frequency represents the exchange's upper limit for flow control, determining the success probability of limit order trades and the level of slippage for market orders. Generally, when the BBO change window is shorter than network latency, the second weight should be increased. The higher the current transaction cost, the greater the third weight.

[0056] Figure 4 This is another flow chart of calculating the transaction network weight in an embodiment of the present invention. In one embodiment, the transaction network weight of the execution node is calculated based on the first weight, the second weight, and the third weight, including:

[0057] Step 401, respectively obtain the average network delay, BBO change frequency, and the corresponding ratio of the current transaction cost;

[0058] Step 402: Calculate the transaction network weight of the execution node by weighted summation based on the first weight, the second weight, the third weight, the average network delay, the BBO change frequency, and the corresponding ratio of the current transaction cost.

[0059] In step 106, the path corresponding to the execution node corresponding to the minimum transaction network weight is determined to be the shortest path. Once the shortest path is determined, the execution node corresponding to the transaction route is obtained.

[0060] In the method proposed in an embodiment of the present invention, a transaction center network is constructed, comprising a decision center node and multiple execution nodes, with the decision center node connected to each execution node via edges, each of which is a path. The average network delay of each execution node is calculated based on its heartbeat data. The BBO change frequency of each execution node is fitted based on its historical BBO data. The current transaction cost of each execution node is fitted based on its historical transaction cost. For each execution node in the transaction center network, the transaction network weight of that execution node is calculated based on its average network delay, BBO change frequency, and current transaction cost. The path corresponding to the execution node with the minimum transaction network weight is determined as the shortest path. Compared to existing technologies, this method uses the average network delay, BBO change frequency, and current transaction cost to determine the transaction network weight of each execution node, thereby determining the shortest path and achieving the most appropriate transaction routing.

[0061] The embodiment of the present invention further provides a transaction route determination device, the principle of which is similar to that of the transaction route determination method, and will not be described in detail here.

[0062] Figure 5 Schematic diagram of a transaction route determination device according to an embodiment of the present invention, comprising:

[0063] The transaction center network construction module 501 is used to construct a transaction center network, wherein the transaction center network includes a decision center node and multiple execution nodes, and the decision center node is connected to each execution node via an edge, wherein the edge is a path;

[0064] The network delay mean calculation module 502 is used to calculate the network delay mean of different execution nodes based on the heartbeat data of different execution nodes;

[0065] The BBO change frequency fitting module 503 is used to fit the BBO change frequency of different execution nodes based on the historical BBO data of different execution nodes;

[0066] The current transaction cost fitting module 504 is used to fit the current transaction costs of different execution nodes based on the historical transaction costs of different execution nodes;

[0067] The transaction network weight calculation module 505 is used to calculate the transaction network weight of each execution node in the transaction center network based on the average network delay, BBO change frequency, and current transaction cost of the execution node;

[0068] The shortest path determination module 506 is configured to determine the path corresponding to the execution node corresponding to the minimum transaction network weight as the shortest path.

[0069] In one embodiment, the network delay mean calculation module is specifically used to:

[0070] The heartbeat data of different execution nodes are processed using the time series method to obtain the heartbeat data sequence;

[0071] The heartbeat data sequence is sent to a stationary model to obtain a network delay mean, where the stationary model is obtained by training historical heartbeat data and historical network delay mean.

[0072] In one embodiment, the transaction network weight calculation module is specifically used to:

[0073] For each execution node, query the correspondence between the average network delay and the transaction cost, obtain the transaction cost corresponding to the average network delay of the execution node, and determine the first weight corresponding to the transaction cost corresponding to the average network delay;

[0074] For each execution node, query the correspondence between BBO change frequency and transaction cost, obtain the transaction cost corresponding to the BBO change frequency of the execution node, and determine the second weight corresponding to the transaction cost corresponding to the BBO change frequency;

[0075] For each execution node, determine the third weight corresponding to the current transaction cost;

[0076] The transaction network weight of the execution node is calculated according to the first weight, the second weight, and the third weight.

[0077] In one embodiment, the transaction network weight calculation module is specifically used to:

[0078] Obtain the average network delay, BBO change frequency, and the corresponding ratio of the current transaction cost respectively;

[0079] The transaction network weight of the execution node is calculated by weighted summation based on the first weight, the second weight, the third weight, the average network delay, the BBO change frequency, and the corresponding ratio of the current transaction cost.

[0080] In summary, the apparatus proposed in the embodiments of the present invention constructs a transaction center network comprising a decision center node and multiple execution nodes, with the decision center node connected to each execution node via edges, each of which is a path. The average network delay of each execution node is calculated based on its heartbeat data. The BBO change frequency of each execution node is fitted based on its historical BBO data. The current transaction cost of each execution node is fitted based on its historical transaction cost. For each execution node in the transaction center network, the transaction network weight of that execution node is calculated based on its average network delay, BBO change frequency, and current transaction cost. The path corresponding to the execution node with the minimum transaction network weight is determined as the shortest path. Compared to existing techniques, this method uses the average network delay, BBO change frequency, and current transaction cost to determine the transaction network weight of each execution node, thereby determining the shortest path and achieving the most appropriate transaction routing.

[0081] An embodiment of the present invention further provides a computer device, Figure 6 Schematic diagram of a computer device in an embodiment of the present invention. The computer device 600 includes a memory 610, a processor 620, and a computer program 630 stored in the memory 610 and executable on the processor 620. When the processor 620 executes the computer program 630, the transaction routing determination method described above is implemented.

[0082] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the method for determining a transaction route is implemented.

[0083] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned transaction route determination method is implemented.

[0084] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program service systems. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program service system implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0085] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program business systems according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as combinations of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0086] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0088] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A transaction routing determination method, characterized in that: include: Constructing a transaction center network, wherein the transaction center network includes a decision center node and multiple execution nodes, and the decision center node is connected to each execution node through an edge, wherein the edge is a path; Calculate the average network delay of different execution nodes based on the heartbeat data of different execution nodes; Based on the historical BBO data of different execution nodes, the BBO change frequency of different execution nodes is obtained by fitting; Based on the historical transaction costs of different execution nodes, the current transaction costs of different execution nodes are obtained by fitting; For each execution node, query the correspondence between the average network delay and the transaction cost, obtain the transaction cost corresponding to the average network delay of the execution node, and determine the first weight corresponding to the transaction cost corresponding to the average network delay; For each execution node, query the correspondence between BBO change frequency and transaction cost, obtain the transaction cost corresponding to the BBO change frequency of the execution node, and determine the second weight corresponding to the transaction cost corresponding to the BBO change frequency; For each execution node, determine the third weight corresponding to the current transaction cost; Obtain the average network delay, BBO change frequency, and the corresponding ratio of the current transaction cost respectively; The transaction network weight of the execution node is calculated by weighted summation based on the first weight, the second weight, the third weight, the average network delay, the BBO change frequency, and the corresponding ratio of the current transaction cost; The path corresponding to the execution node corresponding to the minimum transaction network weight is determined to be the shortest path.

2. The method according to claim 1, wherein Based on the heartbeat data of different execution nodes, the average network delay of different execution nodes is calculated, including: The heartbeat data of different execution nodes are processed using the time series method to obtain the heartbeat data sequence; The heartbeat data sequence is sent to a stationary model to obtain a network delay mean, where the stationary model is obtained by training historical heartbeat data and historical network delay mean.

3. A transaction route determination device, characterized in that: include: A transaction center network construction module is used to construct a transaction center network, wherein the transaction center network includes a decision center node and multiple execution nodes, and the decision center node is connected to each execution node via an edge, wherein the edge is a path; The network delay mean calculation module is used to calculate the network delay mean of different execution nodes based on the heartbeat data of different execution nodes; The BBO change frequency fitting module is used to fit the BBO change frequency of different execution nodes based on the historical BBO data of different execution nodes; The current transaction cost fitting module is used to fit the current transaction costs of different execution nodes based on the historical transaction costs of different execution nodes; The transaction network weight calculation module is used to query the corresponding relationship between the average network delay and the transaction cost for each execution node, obtain the transaction cost corresponding to the average network delay of the execution node, and determine the first weight corresponding to the transaction cost corresponding to the average network delay; query the corresponding relationship between the BBO change frequency and the transaction cost for each execution node, obtain the transaction cost corresponding to the BBO change frequency of the execution node, and determine the second weight corresponding to the transaction cost corresponding to the BBO change frequency; determine the third weight corresponding to the current transaction cost for each execution node; obtain the corresponding ratios of the average network delay, the BBO change frequency, and the current transaction cost respectively; and calculate the transaction network weight of the execution node by weighted summation based on the first weight, the second weight, and the third weight and the corresponding ratios of the average network delay, the BBO change frequency, and the current transaction cost; The shortest path determination module is used to determine the path corresponding to the execution node corresponding to the minimum transaction network weight as the shortest path.

4. The device according to claim 3, characterized in that The network delay mean calculation module is specifically used for: The heartbeat data of different execution nodes are processed using the time series method to obtain the heartbeat data sequence; The heartbeat data sequence is sent to a stationary model to obtain a network delay mean, where the stationary model is obtained by training historical heartbeat data and historical network delay mean.

5. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 2 is implemented.

6. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 2 is implemented.

7. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 2 is implemented.

Images