CN116883176A

CN116883176A - Backtracking test method and device for trend prediction model and electronic equipment

Info

Publication number: CN116883176A
Application number: CN202310847974.2A
Authority: CN
Inventors: 周魁; 皇甫晓洁; 张倩妮; 王航
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2023-07-11
Filing date: 2023-07-11
Publication date: 2023-10-13

Abstract

The disclosure provides a backtracking test method and device for a trend prediction model and electronic equipment, and can be applied to the fields of computers and financial science and technology. The backtracking test method comprises the following steps: obtaining a target model file corresponding to the target trend prediction model; acquiring at least one historical flat transaction record generated in a preset time period, wherein each historical flat transaction record comprises historical transaction time and historical transaction cost; aiming at each historical flat transaction record, taking the historical transaction time corresponding to the historical flat transaction record as a target moment, and acquiring target feature data generated at the target moment according to the target feature information; carrying out trend prediction on the target characteristic data by using a target trend prediction model, and outputting a trend prediction result; performing simulated flat plate transactions according to the trend prediction result, wherein the simulated flat plate transactions comprise simulated transaction costs; and analyzing the prediction accuracy of the target trend prediction model according to the simulated transaction cost and the historical transaction cost.

Description

Backtracking test method and device for trend prediction model and electronic equipment

Technical Field

The present disclosure relates to the field of computers and financial technology, and more particularly to a backtracking test method, apparatus, device, medium and program product for trend prediction models.

Background

In recent years, with the popularity of artificial intelligence applications, various trend prediction models applied to investment transactions are layered. For judging trend prediction models in market making transactions, indexes such as prediction category accuracy, recall rate and the like are generally adopted in the related technology for measurement.

In the process of implementing the inventive concept of the present disclosure, the inventor found that there are at least the following problems in the related art: it is difficult to accurately evaluate the accuracy of the trend prediction model using the related art.

Disclosure of Invention

In view of the foregoing, the present disclosure provides a backtracking test method, apparatus, device, medium, and program product for a trend prediction model.

According to one aspect of the present disclosure, there is provided a backtracking test method for a trend prediction model, including:

responding to a backtracking test request aiming at a target trend prediction model, and acquiring a target model file corresponding to the target trend prediction model from a database, wherein the target model file comprises target characteristic information and a target prediction period;

Obtaining at least one historical flat transaction record generated in a preset time period from a transaction system, wherein each historical flat transaction record comprises historical transaction time and historical transaction cost;

aiming at each historical flat transaction record in the at least one historical flat transaction record, taking the historical transaction time corresponding to the historical flat transaction record as a target moment, and acquiring target characteristic data generated at the target moment according to the target characteristic information;

carrying out trend prediction on the target characteristic data by using the target trend prediction model, and outputting a trend prediction result, wherein the trend prediction result represents a market trend in the target prediction period;

performing a simulated flat plate transaction according to the trend prediction result, wherein the simulated flat plate transaction comprises simulated transaction cost; and

and analyzing the prediction accuracy of the target trend prediction model according to the simulated transaction cost and the historical transaction cost.

According to an embodiment of the present disclosure, the trend prediction result includes one of: rising signal, oscillating signal and falling signal.

According to an embodiment of the present disclosure, the target model file further includes a preset expansion threshold;

Wherein, when the trend prediction result includes an up signal, the performing the simulated flat transaction according to the trend prediction result includes:

monitoring market fluctuation data of each unit moment in the target prediction period;

and under the condition that the market fluctuation data at the current moment exceeds the preset fluctuation threshold, carrying out simulated flat disc transaction at the current moment, wherein the simulated flat disc transaction cost is determined according to the transaction price at the current moment.

According to an embodiment of the present disclosure, the above method further includes:

and under the condition that the market fluctuation data of any unit time in the target prediction period does not exceed the preset fluctuation threshold, carrying out simulated flat disc transaction at the final time of the target prediction period, wherein the simulated flat disc transaction cost is determined according to the transaction price at the final time.

According to an embodiment of the present disclosure, the simulated flat disc transaction is not performed in the case where the trend prediction result includes an oscillation signal.

According to an embodiment of the present disclosure, the target model file includes a preset drop threshold;

wherein, when the trend prediction result includes a drop signal, the performing the simulated flat disc transaction according to the trend prediction result includes:

Monitoring market drop data of each unit moment in the target prediction period;

and under the condition that the market drop data at the current moment exceeds the preset drop threshold, carrying out simulated flat plate transaction at the current moment, wherein the simulated flat plate transaction cost is determined according to the transaction price at the current moment.

and under the condition that the market drop data of any unit time in the target prediction period does not exceed the preset drop threshold, carrying out simulated flat plate transaction at the last time of the target prediction period, wherein the simulated flat plate transaction cost is determined according to the transaction price at the last time.

According to an embodiment of the present disclosure, analyzing the prediction accuracy of the target trend prediction model according to the simulated transaction cost and the historical transaction cost includes:

determining damage and benefit indexes according to a preset formula according to the simulated transaction cost and the historical transaction cost;

and analyzing the prediction accuracy of the target trend prediction model by using the damage index.

According to the embodiment of the disclosure, the history flat transaction records comprise n pieces, wherein n is more than or equal to 2;

The determining the damage index according to the simulated transaction cost and the historical transaction cost and the preset formula comprises the following steps:

aiming at an ith historical flat transaction record in n historical flat transaction records, determining an ith saving cost according to the preset formula according to an ith simulated transaction cost and an ith historical transaction cost corresponding to the ith historical flat transaction record, and finally obtaining n saving costs, wherein i is more than or equal to 1 and less than or equal to n;

and carrying out averaging treatment on the n cost-saving indexes to obtain the damage index.

Another aspect of the present disclosure provides a backtracking test apparatus for a trend prediction model, including:

the first acquisition module is used for responding to a backtracking test request aiming at the target trend prediction model, and acquiring a target model file corresponding to the target trend prediction model from a database, wherein the target model file comprises target characteristic information and a target prediction period;

the second acquisition module is used for acquiring at least one historical flat transaction record generated in a preset time period from the transaction system, wherein each historical flat transaction record comprises historical transaction time and historical transaction cost;

The third acquisition module is used for acquiring target feature data generated at target time according to the target feature information by taking the historical transaction time corresponding to the historical flat transaction record as the target time for each historical flat transaction record in the at least one historical flat transaction record;

the trend prediction module is used for predicting the trend of the target characteristic data by utilizing the target trend prediction model and outputting a trend prediction result, wherein the trend prediction result represents the market trend in the target prediction period;

the simulation module is used for carrying out simulation flat plate transaction according to the trend prediction result, wherein the simulation flat plate transaction comprises simulation transaction cost; and

and the analysis module is used for analyzing the prediction accuracy of the target trend prediction model according to the simulated transaction cost and the historical transaction cost.

Another aspect of the present disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the backtracking test method for the trend prediction model described above.

Another aspect of the present disclosure also provides a computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the backtracking test method described above for a trend prediction model.

Another aspect of the present disclosure also provides a computer program product comprising a computer program which, when executed by a processor, implements the backtracking test method described above for a trend prediction model.

According to the embodiment of the disclosure, the method comprises the steps of acquiring a target model file corresponding to a target trend prediction model and at least one historical flat transaction record generated in a preset time period; taking the historical transaction time corresponding to the historical flat transaction record as a target moment, and acquiring target characteristic data generated at the target moment according to the target characteristic information; then, carrying out trend prediction on the target characteristic data by utilizing a target trend prediction model, and outputting a trend prediction result for representing the market trend in a target prediction period; then carrying out simulated flat plate transaction according to the trend prediction result, wherein the simulated flat plate transaction comprises simulated transaction cost; and then analyzing the prediction accuracy of the target trend prediction model according to the simulated transaction cost and the historical transaction cost, so that the technical problem that the accuracy of the trend prediction model is difficult to accurately evaluate by adopting a related technology is at least partially solved, and the technical effect of accurately evaluating the prediction accuracy of the trend prediction model is achieved.

Drawings

The foregoing and other objects, features and advantages of the disclosure will be more apparent from the following description of embodiments of the disclosure with reference to the accompanying drawings, in which:

FIG. 1 schematically illustrates an application scenario diagram of a backtracking test method, apparatus, device, medium and program product for trend prediction models according to an embodiment of the present disclosure;

FIG. 2 schematically illustrates a flow chart of a backtracking test method for a trend prediction model, according to an embodiment of the present disclosure;

FIG. 3 schematically illustrates a flow chart of a trace-back test method when the trend prediction result is a rising signal, according to an embodiment of the present disclosure;

FIG. 4 schematically illustrates a flow chart of a backtracking test method when the trend prediction result is a falling signal, according to an embodiment of the present disclosure;

FIG. 5 schematically illustrates a block diagram of a backtracking test apparatus for trend prediction models, according to an embodiment of the present disclosure; and

fig. 6 schematically illustrates a block diagram of an electronic device adapted to implement a backtracking test method for trend prediction models, according to an embodiment of the present disclosure.

Detailed Description

Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that the description is only exemplary and is not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the present disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. In addition, in the following description, descriptions of well-known structures and techniques are omitted so as not to unnecessarily obscure the concepts of the present disclosure.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and/or the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.

All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It should be noted that the terms used herein should be construed to have meanings consistent with the context of the present specification and should not be construed in an idealized or overly formal manner.

Where expressions like at least one of "A, B and C, etc. are used, the expressions should generally be interpreted in accordance with the meaning as commonly understood by those skilled in the art (e.g.," a system having at least one of A, B and C "shall include, but not be limited to, a system having a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.).

In market making transactions, the fundamental idea of profitability is that the value of holding a warehouse and the cost of holding the warehouse are higher than the former, the space of profitability is available only, the gap between the two is enlarged, and the key factor is to grasp market trend.

In recent years, with the popularity of artificial intelligence applications, various trend prediction models applied to investment transactions are layered. And for judging the quality of a prediction model, indexes such as prediction category accuracy, recall rate and the like are commonly adopted in the related technology for measurement. However, for the prediction model applied to the transaction, the index is not accurate and direct enough, and it is difficult to accurately judge the trend prediction model. For example, a model predicts a very high win rate, but can only bring less benefit at a time, and once predicted incorrectly, can cause a large penalty. It is clear that they do not intuitively show how much, if applied, the model can bring about a user's profit. Therefore, a real and reasonable backtracking test method is needed for evaluating the advantages and disadvantages of the trend prediction model.

In view of the above, the present disclosure provides a backtracking test method based on real flat-panel transaction to simulate the actual application effect of the trend prediction model. Specifically, the accuracy of the trend prediction model is analyzed by combining the trend prediction result generated by the trend prediction model and the real historical flat transaction record for back measurement. The evaluation mode of the scheme is more visual than indexes such as model accuracy, recall rate and the like, and can be closer to the effect in actual application. According to the evaluation method, a model preferential process is defined, and an actual more effective trend prediction model can be effectively screened out

Specifically, an embodiment of the present disclosure provides a backtracking test method for a trend prediction model, including: responding to a backtracking test request aiming at a target trend prediction model, and acquiring a target model file corresponding to the target trend prediction model from a database, wherein the target model file comprises target characteristic information and a target prediction period; obtaining at least one historical flat transaction record generated in a preset time period from a transaction system, wherein each historical flat transaction record comprises historical transaction time and historical transaction cost; aiming at each historical flat transaction record in the at least one historical flat transaction record, taking the historical transaction time corresponding to the historical flat transaction record as a target moment, and acquiring target characteristic data generated at the target moment according to the target characteristic information; carrying out trend prediction on the target characteristic data by using the target trend prediction model, and outputting a trend prediction result, wherein the trend prediction result represents a market trend in the target prediction period; performing a simulated flat plate transaction according to the trend prediction result, wherein the simulated flat plate transaction comprises simulated transaction cost; and analyzing the prediction accuracy of the target trend prediction model according to the simulated transaction cost and the historical transaction cost.

It should be noted that the backtracking test method and device for the trend prediction model provided by the embodiment of the disclosure can be used in the computer field and the financial field. The backtracking test method and device for the trend prediction model provided by the embodiment of the disclosure can also be used in any field except the computer field and the financial field. The application fields of the backtracking test method and the backtracking test device for the trend prediction model provided by the embodiment of the disclosure are not limited.

In the technical scheme of the disclosure, the authorization or consent of the user is obtained before the personal information of the user is obtained or acquired.

In the technical scheme of the disclosure, the processes of acquiring, collecting, storing, using, processing, transmitting, providing, disclosing, applying and the like of the data all conform to the regulations of related laws and regulations, necessary security measures are adopted, and the public order harmony is not violated.

Fig. 1 schematically illustrates an application scenario diagram of a backtracking test method, apparatus, device, medium and program product for a trend prediction model according to an embodiment of the present disclosure.

As shown in fig. 1, an application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is a medium used to provide a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user may interact with the server 105 through the network 104 using at least one of the first terminal device 101, the second terminal device 102, the third terminal device 103, to receive or send messages, etc. Various communication client applications, such as a financial class application, a shopping class application, a web browser application, a search class application, an instant messaging tool, a mailbox client, social platform software, etc. (by way of example only) may be installed on the first terminal device 101, the second terminal device 102, the third terminal device 103.

The first terminal device 101, the second terminal device 102, the third terminal device 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablets, laptop and desktop computers, and the like.

The server 105 may be a server providing various services, such as a background management server (by way of example only) providing support for websites browsed by the user using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as the user request, and feed back the processing result (e.g., the web page, information, or data obtained or generated according to the user request) to the terminal device.

It should be noted that the backtracking test method for the trend prediction model provided by the embodiments of the present disclosure may be generally performed by the server 105. Accordingly, the backtracking test apparatus for trend prediction models provided by the embodiments of the present disclosure may be generally provided in the server 105. The backtracking test method for the trend prediction model provided by the embodiments of the present disclosure may also be performed by a server or a server cluster that is different from the server 105 and is capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and/or the server 105. Accordingly, the backtracking test apparatus for trend prediction model provided by the embodiments of the present disclosure may also be provided in a server or a server cluster that is different from the server 105 and is capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and/or the server 105.

It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

The backtracking test method for the trend prediction model of the disclosed embodiment will be described in detail below with reference to fig. 2 to 4 based on the scenario described in fig. 1.

Fig. 2 schematically illustrates a flow chart of a backtracking test method for a trend prediction model according to an embodiment of the present disclosure.

As shown in fig. 2, the backtracking test method for a trend prediction model of this embodiment includes operations S210 to S260, which may be performed by a server.

In operation S210, in response to the backtracking test request for the target trend prediction model, a target model file corresponding to the target trend prediction model is acquired from the database, wherein the target model file includes target feature information and a target prediction period.

According to an embodiment of the present disclosure, the target model file is a file when training the target trend prediction model, and is a persistent model file.

According to embodiments of the present disclosure, the target feature information may include feature factors that need to be input to the target trend prediction model. For example, the feature factors may include quotation factors, business bilateral strength comparisons, public opinion factors, and the like.

According to an embodiment of the present disclosure, the target prediction period includes a valid period of a prediction result generated by the target trend prediction model. It should be noted that, the signal generated by the trend prediction model is generally predicted to exist in the effective period. For example, the trend prediction model generates an up signal, which has a specific meaning that the trend prediction model considers that market trends have a certain rise in the future prediction cycle time period. For another example, the trend prediction model generates a drop signal, which specifically means that the trend prediction model considers that the market trend falls to a certain extent in a future prediction period time period.

At operation S220, at least one historical flat transaction record generated within a preset time period is obtained from the transaction system, wherein each historical flat transaction record includes a historical transaction time and a historical transaction cost.

According to embodiments of the present disclosure, the historical flat transaction record may include historical transaction time, historical transaction cost. It should be noted that the historical flat transaction record may also include transaction directions, transaction amounts, transaction prices, and the like. The transaction cost may be determined based on the transaction amount and the transaction price.

Table 1 shows historical transaction detail records according to embodiments of the present disclosure.

TABLE 1

In operation S230, for each of the at least one historical flat transaction record, target feature data generated at a target time is acquired according to the target feature information with a historical transaction time corresponding to the historical flat transaction record as the target time.

According to embodiments of the present disclosure, the target feature data may include quotation data, public opinion data, purchase price difference, momentum factor, etc. generated at the target time.

According to an embodiment of the present disclosure, for example, the target feature information includes feature information a, feature information B, and feature information C. The target feature data may include feature data a corresponding to the feature information a, feature data B corresponding to the feature information B, and feature data C corresponding to the feature information C generated at the target time. More specifically, for example, the target feature information includes a quotation factor, a public opinion factor, and a momentum factor. The target feature data may include generating quotation data, public opinion data, and momentum data at the target time.

According to the embodiment of the disclosure, the target characteristic data is generated by acquiring the target moment, namely the historical transaction time of the historical flat plate transaction, so that the target trend prediction model is conveniently tested back according to the target characteristic data, and the prediction result is compared with the historical flat plate transaction record, so that the accuracy of the target trend prediction model can be more accurately estimated.

In operation S240, trend prediction is performed on the target feature data using the target trend prediction model, and a trend prediction result is output, wherein the trend prediction result characterizes a market trend within the target prediction period.

According to embodiments of the present disclosure, the target prediction period may be set to a minute order, for example, two minutes, three minutes, five minutes, and so on.

In operation S250, a simulated flat plate transaction is performed according to the trend prediction result, wherein the simulated flat plate transaction includes a simulated transaction cost.

According to the embodiment of the disclosure, the moment when the simulated flat disc transaction is carried out can be determined according to the trend prediction result, the simulated flat disc transaction is carried out at the moment, and the simulated transaction cost is the transaction cost at the moment.

In operation S260, the prediction accuracy of the target trend prediction model is analyzed according to the simulated transaction cost and the historical transaction cost.

According to the embodiment of the disclosure, the damage index is determined according to the simulated transaction cost and the historical transaction cost, so that the prediction accuracy of the target trend prediction model is analyzed according to the damage index.

According to the embodiment of the disclosure, the method comprises the steps of acquiring a target model file corresponding to a target trend prediction model and at least one historical flat transaction record generated in a preset time period; taking the historical transaction time corresponding to the historical flat transaction record as a target moment, and acquiring target characteristic data generated at the target moment according to the target characteristic information; then, carrying out trend prediction on the target characteristic data by utilizing a target trend prediction model, and outputting a trend prediction result for representing the market trend in a target prediction period; then carrying out simulated flat plate transaction according to the trend prediction result, wherein the simulated flat plate transaction comprises simulated transaction cost; and then analyzing the prediction accuracy of the target trend prediction model according to the simulated transaction cost and the historical transaction cost, so that the technical problems that the trend prediction model is not accurate and direct enough and is difficult to accurately judge in the related technology by using indexes are at least partially solved, and the technical effects of defining the model preference process and effectively screening out the actual more effective trend prediction model are achieved.

According to an embodiment of the present disclosure, the target model file further includes a preset expansion threshold; wherein, when the trend prediction result includes an up signal, the performing the simulated flat transaction according to the trend prediction result includes: monitoring market fluctuation data of each unit moment in the target prediction period; and under the condition that the market fluctuation data at the current moment exceeds the preset fluctuation threshold, carrying out simulated flat disc transaction at the current moment, wherein the simulated flat disc transaction cost is determined according to the transaction price at the current moment.

According to the embodiment of the disclosure, the preset fluctuation range threshold represents that the market trend is an ascending trend when the fluctuation range exceeds the preset fluctuation range threshold when the fluctuation range is predicted. It should be noted that, when the trend prediction model of the target trend predicts the trend, it is necessary to give a measure of what is calculated, i.e. how much a specific rise is, for example, 1%, and at this time, 1% may be determined as the preset rise threshold.

According to embodiments of the present disclosure, the unit time may be every second or every minute. The monitoring of the market-rising amplitude data at each unit time in the target prediction period may include: and monitoring the market fluctuation data of each second in the target prediction period. Specifically, for example, if the target preset period is 60 seconds and the unit time is every second, monitoring the market expansion data at each unit time in the target prediction period may include monitoring the market expansion data at each second within 60 seconds of the target preset period, so that 60 market expansion data can be obtained.

According to the embodiment of the disclosure, in the process of monitoring the market expansion data at the unit time in the target prediction period, when the market expansion data at the current time exceeds the preset expansion threshold, the simulation transaction can be performed at the current time, and the simulation transaction cost is determined according to the transaction price at the current time.

In one embodiment, for example, the preset rise threshold is 1%, the target prediction period is 60 seconds, and the unit time is every second. When the market fluctuation data generated every second in the target prediction period is monitored for 60 seconds, if the market fluctuation data generated in the 30 th second exceeds the preset fluctuation threshold value by 1%, the simulated flat plate transaction can be performed in the 30 th second, and the simulated transaction cost is determined according to the transaction price in the 30 th second.

According to an embodiment of the present disclosure, the above method further includes: and under the condition that the market fluctuation data of any unit time in the target prediction period does not exceed the preset fluctuation threshold, carrying out simulated flat disc transaction at the final time of the target prediction period, wherein the simulated flat disc transaction cost is determined according to the transaction price at the final time.

According to an embodiment of the present disclosure, in the case where the market fluctuation data at any unit time within the target prediction period does not exceed the preset fluctuation threshold, a compulsory analog flat disc transaction is performed at the final time of the target prediction period.

In one embodiment, for example, the preset rise threshold is 1%, the target prediction period is 60 seconds, and the unit time is every second. When the market fluctuation data generated every second in the 60 seconds of the target prediction period is monitored, if the market fluctuation data in 60 seconds does not exceed the preset fluctuation threshold value by 1%, the simulation flat plate transaction is forced to be carried out in the last moment, namely 60 seconds, and the simulation transaction cost is determined according to the transaction price of 60 seconds.

Fig. 3 schematically illustrates a flowchart of a backtracking test method when the trend prediction result is a rising signal according to an embodiment of the present disclosure.

As shown in fig. 3, the trend prediction result of this embodiment is an up signal, and the backtracking test method thereof includes operations S301 to S304 in addition to operations S210 to S240.

In operation S301, the market-rising amplitude data of each unit moment is monitored in the target prediction period.

In operation S302, it is determined whether or not the market expansion data at the current time exceeds a preset expansion threshold for the market expansion data at each unit time. In case it is determined that the market fluctuation data at the current time exceeds the preset fluctuation threshold, operation S303 is performed; in the case where it is determined that the market expansion data at any unit time within the target prediction period does not exceed the preset expansion threshold, operation S304 is performed.

In operation S303, a simulated flat disc transaction is performed at the current time, wherein a simulated flat disc transaction cost is determined according to a transaction price at the current time.

In operation S304, a simulated flat disc transaction is performed at a final time of the target prediction period, wherein a simulated flat disc transaction cost is determined according to a transaction price at the final time.

According to embodiments of the present disclosure, when the trend prediction result is an oscillating signal, no simulated flat disc transactions are performed, i.e., the data does not participate in the evaluation of the accuracy of the target trend prediction model.

According to an embodiment of the present disclosure, the target model file includes a preset drop threshold; wherein, when the trend prediction result includes a drop signal, the performing the simulated flat disc transaction according to the trend prediction result includes: monitoring market drop data of each unit moment in the target prediction period; and under the condition that the market drop data at the current moment exceeds the preset drop threshold, carrying out simulated flat plate transaction at the current moment, wherein the simulated flat plate transaction cost is determined according to the transaction price at the current moment.

According to the embodiment of the disclosure, the preset drop threshold represents that the market trend is a falling trend when the drop exceeds the preset drop threshold when the rising and falling are predicted. It should be noted that, similar to the preset fluctuation range threshold, when the trend prediction model performs trend prediction, a measure of what to calculate the drop must be given, i.e. how much the specific drop is, for example, 1%, and at this time, 1% may be determined as the preset drop range threshold.

According to an embodiment of the present disclosure, monitoring the market fall data of each unit time in the target prediction period may include: and monitoring the market drop data of each second in the target prediction period. Specifically, for example, the target preset period is 5 minutes, and the unit time is every minute, and monitoring the market drop data of each unit time in the target prediction period may include monitoring the market drop data of every minute in the target preset period for 5 minutes, so that 5 market drop data can be obtained.

According to the embodiment of the disclosure, in the process of monitoring the market drop data at the unit moment in the target prediction period, under the condition that the market drop data at the current moment exceeds the preset drop threshold, the simulation transaction can be performed at the current moment, and the simulation transaction cost is determined according to the transaction price at the current moment.

In one embodiment, for example, the preset drop threshold is 1%, the target prediction period is 5 minutes, and the unit time is every minute. When the market fluctuation data generated every minute in the 5-minute period of the target prediction is monitored, if the market fluctuation data generated in the 3 rd minute exceeds the preset fluctuation threshold value by 1%, the simulated flat plate transaction can be performed in the 3 rd minute, and the simulated transaction cost is determined according to the transaction price in the 3 rd minute.

According to an embodiment of the present disclosure, the above method further includes: and under the condition that the market drop data of any unit time in the target prediction period does not exceed the preset drop threshold, carrying out simulated flat plate transaction at the last time of the target prediction period, wherein the simulated flat plate transaction cost is determined according to the transaction price at the last time.

According to an embodiment of the present disclosure, in a case where the market fall data at any unit time within the target prediction period does not exceed the preset fall threshold, a compulsory simulated flat plate transaction is performed at the last time of the target prediction period.

In one embodiment, for example, the preset drop threshold is 1%, the target prediction period is 5 minutes, and the unit time is every minute. And when the market drop data generated every minute in the 5-minute period of the target prediction is monitored, if the market drop data in 5-minute period does not exceed the preset drop threshold value by 1%, forcing the simulation flat plate transaction to be carried out at the last moment, namely the 5-minute period, and determining the simulation transaction cost according to the transaction price of the 5-minute period.

Fig. 4 schematically illustrates a flowchart of a backtracking test method when the trend prediction result is a falling signal according to an embodiment of the present disclosure.

As shown in fig. 4, the trend prediction result of this embodiment is a falling signal, and the backtracking test method thereof includes operations S401 to S404 in addition to operations S210 to S240.

In operation S401, the market drop data of each unit moment is monitored in the target prediction period.

In operation S402, it is determined, for each unit time, whether the current time of the market drop data exceeds a preset drop threshold. Executing operation S403 when it is determined that the market drop data at the current time exceeds the preset drop threshold; if it is determined that the market drop data at any unit time in the target prediction period does not exceed the preset drop threshold, operation S404 is executed.

In operation S403, a simulated flat disc transaction is performed at the current time, wherein the simulated flat disc transaction cost is determined according to the transaction price at the current time.

In operation S404, a simulated flat disc transaction is performed at a last moment of the target prediction period, wherein a simulated flat disc transaction cost is determined according to a transaction price at the last moment.

According to an embodiment of the present disclosure, analyzing the prediction accuracy of the target trend prediction model according to the simulated transaction cost and the historical transaction cost includes: determining damage and benefit indexes according to a preset formula according to the simulated transaction cost and the historical transaction cost; and analyzing the prediction accuracy of the target trend prediction model by using the damage index.

According to an embodiment of the present disclosure, determining the damage index according to the predetermined formula from the simulated transaction cost and the historical transaction cost may include: and subtracting the simulated transaction cost and the historical transaction cost for each historical flat transaction record in at least one historical flat transaction record to obtain the cost saving corresponding to the historical transaction record.

According to the embodiment of the disclosure, the history flat transaction records comprise n pieces, wherein n is more than or equal to 2; the determining the damage index according to the simulated transaction cost and the historical transaction cost and the preset formula comprises the following steps: aiming at an ith historical flat transaction record in n historical flat transaction records, determining an ith saving cost according to the preset formula according to an ith simulated transaction cost and an ith historical transaction cost corresponding to the ith historical flat transaction record, and finally obtaining n saving costs, wherein i is more than or equal to 1 and less than or equal to n; and carrying out averaging treatment on the n cost-saving indexes to obtain the damage index.

According to an embodiment of the present disclosure, when the historical flat-disk transaction record includes n, a comparison of the price of the transaction is made for each generated simulated transaction, and the average flat-disk cost saved per transaction is calculated. At this time, the preset formula may be as shown in formula (1).

Where n represents the analog transaction number, size_cost _i Representing the simulated transaction price of the ith pen, original_cost _i For the i-th historical transaction price qty _i Transaction amount of the ith pen.

In one embodiment, assuming that the prediction period of the target trend prediction model is 5 minutes, if the trend prediction result in the preset period is that the rising signal indicates that the rising amplitude exceeds 1% in 5 minutes, the trend prediction result is that the falling signal indicates that the falling amplitude exceeds 1% in 5 minutes. At this time, backtracking test was performed on three mermaid coin on-demand flat plate transaction details, and the results are shown in table 2.

TABLE 2

/>

As can be seen from table 2, the cost saving is averaged 22967, i.e. the target trend prediction model returns the average cost saving 22967 yuan of people and civil coins per flat plate transaction detail. Therefore, the embodiment of the disclosure carries out the return test by combining the signal generated by the model and the real flat disc transaction, and carries out the damage calculation according to the proposed formula, and the evaluation mode is more visual than the indexes such as the model accuracy, the recall rate and the like, and can be closer to the effect in actual application. According to the evaluation method, a model preferential process is defined, and an actual more effective trend prediction model can be effectively screened out.

It should be noted that, unless there is an execution sequence between different operations or an execution sequence between different operations in technical implementation, the execution sequence between multiple operations may be different, and multiple operations may also be executed simultaneously in the embodiment of the disclosure.

Based on the backtracking test method aiming at the trend prediction model, the disclosure also provides a backtracking test device aiming at the trend prediction model. The device will be described in detail below in connection with fig. 5.

Fig. 5 schematically illustrates a block diagram of a backtracking test apparatus for trend prediction models according to an embodiment of the present disclosure.

As shown in fig. 5, the backtracking test apparatus 500 for a trend prediction model of this embodiment includes a first acquisition module 510, a second acquisition module 520, a third acquisition module 530, a trend prediction module 540, a simulation module 550, and an analysis module 560.

The first obtaining module 510 is configured to obtain, from a database, a target model file corresponding to the target trend prediction model in response to a backtracking test request for the target trend prediction model, where the target model file includes target feature information and a target prediction period. In an embodiment, the first obtaining module 510 may be configured to perform the operation S210 described above, which is not described herein.

The second obtaining module 520 is configured to obtain, from the transaction system, at least one historical flat transaction record generated during a preset time period, where each historical flat transaction record includes a historical transaction time and a historical transaction cost. In an embodiment, the second obtaining module 520 may be configured to perform the operation S220 described above, which is not described herein.

The third obtaining module 530 is configured to obtain, for each of the at least one historical flat transaction record, target feature data generated at a target time according to target feature information, with a historical transaction time corresponding to the historical flat transaction record as the target time. In an embodiment, the third obtaining module 530 may be configured to perform the operation S230 described above, which is not described herein.

The trend prediction module 540 is configured to perform trend prediction on the target feature data using a target trend prediction model, and output a trend prediction result, where the trend prediction result characterizes a market trend in a target prediction period. In an embodiment, the trend prediction module 540 may be used to perform the operation S240 described above, which is not described herein.

The simulation module 550 is configured to perform a simulated flat disc transaction according to the trend prediction result, where the simulated flat disc transaction includes a simulated transaction cost. In an embodiment, the simulation module 550 may be used to perform the operation S250 described above, which is not described herein.

The analysis module 560 is configured to analyze the prediction accuracy of the target trend prediction model based on the simulated transaction costs and the historical transaction costs. In an embodiment, the analysis module 560 may be configured to perform the operation S260 described above, which is not described herein.

According to an embodiment of the present disclosure, the target model file further includes a preset expansion threshold.

According to an embodiment of the present disclosure, in the case that the trend prediction result includes an up signal, the simulation module further includes a first monitoring sub-module and a first simulation sub-module.

And the first monitoring sub-module is used for monitoring the market fluctuation data of each unit moment in the target prediction period.

And the first simulation sub-module is used for carrying out simulation flat-disc transaction at the current moment under the condition that the market fluctuation data at the current moment exceeds the preset fluctuation threshold, wherein the simulation flat-disc transaction cost is determined according to the transaction price at the current moment.

According to an embodiment of the present disclosure, the above-described simulation module further includes: and a second analog sub-module.

And the second simulation sub-module is used for carrying out simulation flat plate transaction at the final moment of the target prediction period under the condition that the market fluctuation data of any unit moment in the target prediction period does not exceed the preset fluctuation threshold value, wherein the simulation flat plate transaction cost is determined according to the transaction price at the final moment.

According to an embodiment of the present disclosure, the object model file includes a preset drop threshold.

According to an embodiment of the present disclosure, in a case where the trend prediction result includes a falling signal, the simulation module further includes: a second monitoring sub-module and a third simulation sub-module.

And the second monitoring sub-module is used for monitoring the market drop data of each unit moment in the target prediction period.

And the third simulation sub-module is used for carrying out simulation flat plate transaction at the current moment under the condition that the market drop data at the current moment exceeds the preset drop threshold, wherein the simulation flat plate transaction cost is determined according to the transaction price at the current moment.

According to an embodiment of the present disclosure, the above-described simulation module further includes: and a fourth analog sub-module.

And the fourth simulation sub-module is used for carrying out simulation flat plate transaction at the final moment of the target prediction period under the condition that the market drop data of any unit moment in the target prediction period is not beyond the preset drop threshold value, wherein the simulation flat plate transaction cost is determined according to the transaction price at the final moment.

According to an embodiment of the present disclosure, the analysis module includes: a determination sub-module and an analysis sub-module.

And the determining submodule is used for determining the damage index according to a preset formula according to the simulated transaction cost and the historical transaction cost.

And the analysis sub-module is used for analyzing the prediction accuracy of the target trend prediction model by using the damage index.

According to an embodiment of the present disclosure, the history flat transaction record includes n pieces, where n is equal to or greater than 2.

According to an embodiment of the present disclosure, the determining submodule includes: a determining unit and an averaging unit.

The determining unit is used for determining the ith cost saving according to the preset formula according to the ith simulated transaction cost and the ith historical transaction cost corresponding to the ith historical flat transaction record in the n historical flat transaction records, and finally obtaining n cost saving, wherein i is more than or equal to 1 and less than or equal to n.

And the averaging unit is used for carrying out averaging processing on the n cost-saving indexes to obtain the damage index.

Any number of modules, sub-modules, units, sub-units, or at least some of the functionality of any number of the sub-units according to embodiments of the present disclosure may be implemented in one module. Any one or more of the modules, sub-modules, units, sub-units according to embodiments of the present disclosure may be implemented as split into multiple modules. Any one or more of the modules, sub-modules, units, sub-units according to embodiments of the present disclosure may be implemented at least in part as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system-on-chip, a system-on-substrate, a system-on-package, an Application Specific Integrated Circuit (ASIC), or in any other reasonable manner of hardware or firmware that integrates or encapsulates the circuit, or in any one of or a suitable combination of three of software, hardware, and firmware. Alternatively, one or more of the modules, sub-modules, units, sub-units according to embodiments of the present disclosure may be at least partially implemented as computer program modules, which when executed, may perform the corresponding functions.

According to an embodiment of the present disclosure, any of the first acquisition module 510, the second acquisition module 520, the third acquisition module 530, the trend prediction module 540, the simulation module 550, and the analysis module 560 may be combined in one module to be implemented, or any of the modules may be split into a plurality of modules. Alternatively, at least some of the functionality of one or more of the modules may be combined with at least some of the functionality of other modules and implemented in one module. According to embodiments of the present disclosure, at least one of the first acquisition module 510, the second acquisition module 520, the third acquisition module 530, the trend prediction module 540, the simulation module 550, and the analysis module 560 may be implemented at least in part as hardware circuitry, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system-on-chip, a system-on-substrate, a system-on-package, an Application Specific Integrated Circuit (ASIC), or as hardware or firmware in any other reasonable manner of integrating or packaging the circuitry, or as any one of or a suitable combination of any of the three. Alternatively, at least one of the first acquisition module 510, the second acquisition module 520, the third acquisition module 530, the trend prediction module 540, the simulation module 550, and the analysis module 560 may be at least partially implemented as a computer program module, which may perform the corresponding functions when being executed.

It should be noted that, in the embodiment of the present disclosure, the part of the backtracking test device for the trend prediction model corresponds to the part of the backtracking test method for the trend prediction model in the embodiment of the present disclosure, and the description of the part of the backtracking test device for the trend prediction model specifically refers to the part of the backtracking test method for the trend prediction model, which is not described herein again.

As shown in fig. 6, an electronic device 600 according to an embodiment of the present disclosure includes a processor 601 that can perform various appropriate actions and processes according to a program stored in a read only memory (RON) 602 or a program loaded from a storage section 608 into a random access memory (RAN) 603. The processor 601 may include, for example, a general purpose microprocessor (e.g., a CPU), an instruction set processor and/or an associated chipset and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), or the like. Processor 601 may also include on-board memory for caching purposes. The processor 601 may comprise a single processing unit or a plurality of processing units for performing different actions of the method flows according to embodiments of the disclosure.

In the RAN 603, various programs and data required for the operation of the electronic device 600 are stored. The processor 601, RON 602, and RAN 603 are connected to each other by a bus 604. The processor 601 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in the RON 602 and/or the RAN 603. It should be noted that the program may also be stored in one or more memories other than RON 602 and RAN 603. The processor 601 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in the one or more memories.

According to an embodiment of the present disclosure, the electronic device 600 may also include an input/output (I/O) interface 605, the input/output (I/O) interface 605 also being connected to the bus 604. The electronic device 600 may also include one or more of the following components connected to the I/O interface 605: an input portion 606 including a keyboard, mouse, etc.; an output portion 607 including a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, a speaker, and the like; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. The drive 610 is also connected to the I/O interface 605 as needed. Removable media 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is installed as needed on drive 610 so that a computer program read therefrom is installed as needed into storage section 608.

The present disclosure also provides a computer-readable storage medium that may be embodied in the apparatus/device/system described in the above embodiments; or may exist alone without being assembled into the apparatus/device/system. The computer-readable storage medium carries one or more programs which, when executed, implement methods in accordance with embodiments of the present disclosure.

According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example, but is not limited to: portable computer magnetic disks, hard disks, random access memory (RAN), read-only memory (RON), erasable programmable read-only memory (EPRON or flash memory), portable compact disc read-only memory (CD-RON), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing. In the context of this disclosure, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. For example, according to embodiments of the present disclosure, the computer-readable storage medium may include RON 602 and/or RAN 603 and/or one or more memories other than RON 602 and RAN 603 described above.

Embodiments of the present disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowcharts. The program code, when executed in a computer system, is configured to cause the computer system to implement the retrospective testing method for trend predictive models provided by embodiments of the present disclosure.

The above-described functions defined in the system/apparatus of the embodiments of the present disclosure are performed when the computer program is executed by the processor 601. The systems, apparatus, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the disclosure.

In one embodiment, the computer program may be based on a tangible storage medium such as an optical storage device, a magnetic storage device, or the like. In another embodiment, the computer program may also be transmitted, distributed in the form of signals over a network medium, and downloaded and installed via the communication section 609, and/or installed from the removable medium 611. The computer program may include program code that may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.

In such an embodiment, the computer program may be downloaded and installed from a network through the communication portion 609, and/or installed from the removable medium 611. The above-described functions defined in the system of the embodiments of the present disclosure are performed when the computer program is executed by the processor 601. The systems, devices, apparatus, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the disclosure.

According to embodiments of the present disclosure, program code for performing computer programs provided by embodiments of the present disclosure may be written in any combination of one or more programming languages, and in particular, such computer programs may be implemented in high-level procedural and/or object-oriented programming languages, and/or assembly/machine languages. Programming languages include, but are not limited to, such as Java, c++, python, "C" or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., connected via the Internet using an Internet service provider).

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Those skilled in the art will appreciate that the features recited in the various embodiments of the disclosure and/or in the claims may be provided in a variety of combinations and/or combinations, even if such combinations or combinations are not explicitly recited in the disclosure. In particular, the features recited in the various embodiments of the present disclosure and/or the claims may be variously combined and/or combined without departing from the spirit and teachings of the present disclosure. All such combinations and/or combinations fall within the scope of the present disclosure.

The embodiments of the present disclosure are described above. However, these examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the embodiments cannot be used advantageously in combination. The scope of the disclosure is defined by the appended claims and equivalents thereof. Various alternatives and modifications can be made by those skilled in the art without departing from the scope of the disclosure, and such alternatives and modifications are intended to fall within the scope of the disclosure.

Claims

1. A backtracking test method for a trend prediction model comprises the following steps:

obtaining at least one historical flat transaction record generated within a preset time period from a transaction system, wherein each historical flat transaction record comprises historical transaction time and historical transaction cost;

aiming at each historical flat transaction record in the at least one historical flat transaction record, taking the historical transaction time corresponding to the historical flat transaction record as a target moment, and acquiring target feature data generated at the target moment according to the target feature information;

Carrying out trend prediction on the target feature data by utilizing the target trend prediction model, and outputting a trend prediction result, wherein the trend prediction result represents a market trend in the target prediction period;

performing simulated flat plate transactions according to the trend prediction results, wherein the simulated flat plate transactions comprise simulated transaction costs; and

2. The method of claim 1, wherein the trend prediction result comprises one of: rising signal, oscillating signal and falling signal.

3. The method of claim 2, wherein the object model file further comprises a preset rise threshold;

wherein, in the case that the trend prediction result includes an up signal, the performing the simulated flat disc transaction according to the trend prediction result includes:

and under the condition that the market fluctuation data at the current moment exceeds the preset fluctuation threshold, carrying out simulated flat plate transaction at the current moment, wherein the simulated flat plate transaction cost is determined according to the transaction price at the current moment.

4. A method according to claim 3, further comprising:

and under the condition that the market fluctuation data of any unit time in the target prediction period does not exceed the preset fluctuation threshold, carrying out simulated flat plate transaction at the last time of the target prediction period, wherein the simulated flat plate transaction cost is determined according to the transaction price at the last time.

5. The method of claim 1, wherein the simulated flat disc transaction is not conducted in the event that the trend prediction result includes an oscillating signal.

6. The method of claim 2, wherein the object model file includes a preset fall threshold;

wherein, in the case that the trend prediction result includes a drop signal, the performing the simulated flat disc transaction according to the trend prediction result includes:

monitoring the market drop data of each unit moment in the target prediction period;

7. The method of claim 6, further comprising:

8. The method of claim 1, wherein said analyzing the predictive accuracy of the target trend predictive model from the simulated transaction cost and the historical transaction cost comprises:

9. The method of claim 8, wherein the historical flat-disk transaction record comprises n pieces, wherein n is ≡2;

10. A backtracking test device for a trend prediction model, comprising:

the first acquisition module is used for responding to a backtracking test request aiming at a target trend prediction model, and acquiring a target model file corresponding to the target trend prediction model from a database, wherein the target model file comprises target characteristic information and a target prediction period;

11. An electronic device, comprising:

one or more processors;

storage means for storing one or more programs,

wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of claims 1-9.

12. A computer readable storage medium having stored thereon executable instructions which, when executed by a processor, cause the processor to perform the method according to any of claims 1 to 9.

13. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 9.