CN113362185A

CN113362185A - Investment portfolio data processing method and device

Info

Publication number: CN113362185A
Application number: CN202110629704.5A
Authority: CN
Inventors: 张羽方; 冯程; 杨超; 嵇海锋
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2021-06-07
Filing date: 2021-06-07
Publication date: 2021-09-07

Abstract

The embodiment of the application provides a method and a device for processing investment portfolio data, which can be used in the technical field of artificial intelligence, wherein the method comprises the following steps: screening among a plurality of preset objective functions and a plurality of preset constraint conditions according to the target investment requirement information to form an investment portfolio strategy selection model comprising the objective functions and at least one constraint condition; and taking the investment data of each target investment object as a training sample, training the investment portfolio strategy selection model based on a preset multi-target particle swarm algorithm, obtaining a plurality of target solutions of the investment portfolio strategy selection model for representing the investment portfolio strategy, and generating investment portfolio strategy recommendation result data containing each target solution. The method and the device can effectively improve the efficiency and the intelligent degree of the investment portfolio data processing process and can effectively improve the accuracy and the reliability of the investment portfolio recommendation result.

Description

Investment portfolio data processing method and device

Technical Field

The application relates to the technical field of data processing, in particular to the technical field of artificial intelligence, and specifically relates to a method and a device for processing investment portfolio data.

Background

With the rapid development of the financial industry, the investment types and the amount become more and more diversified, and the current huge investment system is not limited to single investment but needs more investment combinations for users such as trading staff to select. In the process of selecting the investment portfolio, multiple factors such as income, risk control, cost and the like need to be considered, so that the investment portfolio data needs to be effectively processed to ensure the reliability of the investment portfolio result.

At present, the existing investment portfolio selection process usually needs too many people to participate, the investment objects are selected according to the business knowledge and the trading experience of traders, and the historical performance of various open market investments is analyzed according to the business knowledge to predict the future possible profitability; however, when a financial product needs to take multiple objectives and constraints such as income, risk control, cost and the like into consideration, a certain deviation may exist in the position taken by the investment portfolio selected by a trader, so that a more optimal selection cannot be achieved, the fluctuation rate caused by uncertain factors is high, time and labor are consumed in manual processing, and the like, so that the existing investment portfolio selection process has the problems of low accuracy, low intelligence degree, poor efficiency and the like.

Disclosure of Invention

Aiming at the problems in the prior art, the application provides a method and a device for processing investment portfolio data, which can effectively improve the efficiency and the intelligent degree of the investment portfolio data processing process and can effectively improve the accuracy and the reliability of the investment portfolio recommendation result.

In order to solve the technical problem, the application provides the following technical scheme:

in a first aspect, the present application provides a portfolio data processing method, comprising:

screening among a plurality of preset objective functions and a plurality of preset constraint conditions according to the target investment requirement information to form an investment portfolio strategy selection model comprising the objective functions and at least one constraint condition;

and taking the investment data of each target investment object as a training sample, training the investment portfolio strategy selection model based on a preset multi-target particle swarm algorithm, obtaining a plurality of target solutions of the investment portfolio strategy selection model for representing the investment portfolio strategy, and generating investment portfolio strategy recommendation result data containing each target solution.

Further, the training of the investment portfolio strategy selection model based on a preset multi-target particle swarm algorithm with the respective investment data of each target investment object as a training sample to obtain a plurality of target solutions of the investment portfolio strategy selection model for representing the investment portfolio strategy comprises:

Generating an initial population corresponding to the multi-target particle swarm algorithm according to each target investment object;

initializing a preset external file according to the initial population;

a speed position updating step: updating the speed and the position of the particles in the external file to form a new particle population, and storing the new particle population into a preset temporary population;

forming a mixed population according to the temporary population and the initialization population;

acquiring the current global optimal position and the current individual optimal position of the mixed population;

adding 1 to the current iteration number, judging whether the new iteration number is larger than the preset training cycle number, if so, outputting the current global optimal position and the current individual optimal position of the mixed population to be used as the current target solution of the investment portfolio strategy selection model;

and if the new iteration number is less than or equal to the training loop times, returning to execute the speed position updating step.

Further, the generating an initial population corresponding to a multi-target particle swarm algorithm according to each target investment object includes:

setting a plurality of particles according to the target investment objects, wherein each particle is used for representing a unique investment scheme of a uniquely corresponding target investment object within a preset time range;

Combining the particles to form a particle population containing all investment strategy combinations corresponding to the target investment objects;

and carrying out initialization processing on the particle population to obtain an initial population.

Further, initializing a preset external profile according to the initial population includes:

determining an objective function value of each particle in the initial population based on the investment portfolio strategy selection model, and determining a fitness function value of each particle in the initial population based on a preset particle swarm fitness function;

and storing the particles with the fitness function value smaller than 0 into a preset external archive, selecting an individual optimal position and a global optimal position in the sequence from small fitness function values to large fitness function values of the particles in the external archive, and setting the current iteration number as 0.

Further, the forming a mixed population according to the temporary population and the initialization population includes:

determining a fitness function value of each particle in the temporary population based on the particle swarm fitness function;

and mixing the temporary population and the initialization population to form a mixed population.

Further, the acquiring the current global optimal position and the current individual optimal position of the mixed population includes:

performing championship selection processing on the mixed population to generate a global external file and an individual external file corresponding to the mixed population;

and selecting the current global optimal position of the mixed population from the global external files, and selecting the current individual optimal position of the mixed population from the individual external files.

Further, before the screening among a plurality of preset objective functions and a plurality of preset constraints according to the target investment requirement information to form an investment portfolio strategy selection model including the plurality of objective functions and at least one constraint, the method further includes:

receiving an investment portfolio strategy recommendation request, wherein the investment portfolio strategy recommendation request comprises target investment requirement information and a plurality of specified target investment objects;

correspondingly, after generating the investment portfolio strategy recommendation result data containing each target solution, the method further comprises the following steps:

and sending the investment portfolio strategy recommendation result data to the sender of the investment portfolio strategy recommendation request.

In a second aspect, the present application provides a portfolio data processing apparatus comprising:

the model selection module is used for screening a plurality of preset objective functions and a plurality of preset constraint conditions according to the target investment requirement information to form an investment portfolio strategy selection model comprising the objective functions and at least one constraint condition;

and the strategy recommendation module is used for taking the respective investment data of each target investment object as a training sample, training the investment portfolio strategy selection model based on a preset multi-target particle swarm algorithm, obtaining a plurality of target solutions of the investment portfolio strategy selection model, wherein the target solutions are used for representing the investment portfolio strategy, and generating investment portfolio strategy recommendation result data containing each target solution.

In a third aspect, the present application provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the portfolio data processing method when executing the program.

In a fourth aspect, the present application provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the portfolio data processing method.

According to the technical scheme, the investment portfolio data processing method and device provided by the application comprise the following steps: screening among a plurality of preset objective functions and a plurality of preset constraint conditions according to the target investment requirement information to form an investment portfolio strategy selection model comprising the objective functions and at least one constraint condition; the method comprises the steps of taking the respective investment data of each target investment object as a training sample, training the investment portfolio strategy selection model based on a preset multi-target particle swarm algorithm to obtain a plurality of target solutions of the investment portfolio strategy selection model for representing an investment portfolio strategy, generating investment portfolio strategy recommendation result data containing each target solution, obtaining the investment portfolio strategy recommendation result by adopting the investment portfolio strategy selection model containing a plurality of target functions and at least one constraint condition, effectively improving the efficiency and the intelligent degree of the investment portfolio data processing process, effectively improving the accuracy and the reliability of the investment portfolio recommendation result, solving the problems of high dependence of human factors, poor selection result, high fluctuation rate and the like of the existing method, and further providing reliable and stable trading decisions for users such as traders and the like, the profitability of the asset management business is effectively improved, and the investment risk is reduced, so that the user experience is improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a schematic diagram of interaction between a portfolio data processing apparatus and a client device in an embodiment of the present application.

Fig. 2 is a first flowchart of a portfolio data processing method in an embodiment of the present application.

Fig. 3 is a schematic flow chart of step 200 in the portfolio data processing method in an embodiment of the present application.

Fig. 4 is a schematic flow chart of step 210 in the portfolio data processing method in an embodiment of the present application.

Fig. 5 is a schematic flow chart of step 220 in the portfolio data processing method in an embodiment of the present application.

Fig. 6 is a flowchart illustrating step 240 of the portfolio data processing method in an embodiment of the present application.

Fig. 7 is a schematic flow chart of step 250 in the portfolio data processing method in the present embodiment.

Fig. 8 is a second flowchart of the portfolio data processing method in the example of the present application.

Fig. 9 is a schematic configuration diagram of a portfolio data processing apparatus in an embodiment of the present application.

Fig. 10 is a schematic flow chart of a method for processing portfolio data in an example of the application of the present application.

Fig. 11 is a schematic structural diagram of an electronic device in an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

It should be noted that the investment portfolio data processing method and apparatus disclosed in the present application can be used in the technical field of artificial intelligence, and can also be used in any field except the technical field of artificial intelligence.

Taking the bank asset management business field as an example, when a trader builds an investment combination for a collected fund pool, the trader mainly selects investment products according to business knowledge and trading experience of the trader, the trader predicts future possible profitability by analyzing historical performance of various open market investment products according to the business knowledge of the trader, and then selects investment products of a proper investment combination to take a position by combining the goals of risk avoidance and the like.

The existing investment portfolio selection strategy in the field of bank asset management business greatly depends on the business capability and experience accumulation of traders, when the multiple targets and constraints such as income, risk control, cost and the like are considered in the face of financial management products, certain deviation may exist in the position taken by the investment portfolio selected by the traders, the optimal selection cannot be achieved, and the fluctuation rate caused by uncertain factors is high, so that the investment portfolio data processing method based on the artificial intelligence algorithm is very important.

Based on this, embodiments of the present application respectively provide an investment portfolio data processing method, an investment portfolio data processing apparatus, and an electronic device computer readable storage medium, which perform a screening among a plurality of preset objective functions and a plurality of preset constraint conditions according to target investment requirement information to form an investment portfolio strategy selection model including the plurality of objective functions and at least one constraint condition; the method comprises the steps of taking the respective investment data of each target investment object as a training sample, training the investment portfolio strategy selection model based on a preset multi-target particle swarm algorithm to obtain a plurality of target solutions of the investment portfolio strategy selection model for representing an investment portfolio strategy, generating investment portfolio strategy recommendation result data containing each target solution, obtaining the investment portfolio strategy recommendation result by adopting the investment portfolio strategy selection model containing a plurality of target functions and at least one constraint condition, effectively improving the efficiency and the intelligent degree of the investment portfolio data processing process, effectively improving the accuracy and the reliability of the investment portfolio recommendation result, solving the problems of high dependence of human factors, poor selection result, high fluctuation rate and the like of the existing method, and further providing reliable and stable trading decisions for users such as traders and the like, the profitability of the asset management business is effectively improved, and the investment risk is reduced, so that the user experience is improved.

In one or more embodiments of the present application, the portfolio is a set of a plurality of investment items, and after the recruitment period of the financial product, the trader will take the total funds recruited to purchase different types of investment items, such as bonds, stocks, etc., for earning income, and the portfolio purchased by the fund recruited by the financial product is the portfolio.

In one or more embodiments of the present application, the multi-objective optimization problem refers to that an objective to be optimized is not a common single objective, but is composed of multiple mutually exclusive objectives, and therefore needs to be solved by using a specific multi-objective optimization algorithm, and a solution set is different from a common single-objective optimization problem, and the multi-objective optimization problem generally has no unique solution, but a set of non-inferior solution sets. When the multi-objective optimization problem is solved, a plurality of constraint conditions for the parameters are provided, and the problem meeting the conditions is called the multi-objective multi-optimization problem.

In one or more embodiments of the present application, the particle swarm algorithm is an artificial intelligence algorithm, and the inspiration of the application examples of the particle swarm algorithm is derived from the activities of bird groups, and the particle swarm algorithm is widely welcomed by the advantages of simple concept, simple parameter setting, independence of optimization results and initial values, parallelism and the like. The multi-target particle swarm algorithm is to solve the multi-target problem by adopting an evolutionary algorithm idea on the basis of the particle swarm algorithm.

In one or more embodiments of the present application, the fitness function refers to: in the multi-objective particle swarm optimization, an evaluation function is needed to evaluate the advantages and disadvantages of the particles after each evolution and updating, and the fitness function has the effect that in the multi-objective optimization problem, a single and simple objective function cannot be selected as the fitness function due to mutual balance among all objectives, and the calculation of the generally selected fitness function has certain complexity.

Based on the above, the present application further provides a portfolio data processing apparatus for implementing the portfolio data processing method provided in one or more embodiments of the present application, and referring to fig. 1, the portfolio data processing apparatus may be in communication connection with a client device owned by a user by itself or through a third party server, etc., the portfolio data processing apparatus may be a server, receive a portfolio policy recommendation request sent by the user from the client device, or obtain a relevant configuration file preset by the user locally from the client device, a third party database, or the client device, such as at least one of the multiple objective functions, multiple constraint conditions, multiple objective particle swarm algorithms, particle swarm fitness functions, tournament selection processing methods, etc. mentioned in one or more embodiments of the present application. After generating the investment portfolio strategy recommendation result data including each target solution, the investment portfolio data processing device can also send the investment portfolio strategy recommendation result data to the client equipment sending the investment portfolio strategy recommendation request for display, so that a user can timely obtain the investment portfolio strategy recommendation result and the like.

It is understood that the client devices may include smart phones, tablet electronic devices, network set-top boxes, portable computers, desktop computers, Personal Digital Assistants (PDAs), in-vehicle devices, smart wearable devices, and the like. Wherein, intelligence wearing equipment can include intelligent glasses, intelligent wrist-watch, intelligent bracelet etc..

The client device may have a communication module (i.e., a communication unit), and may be communicatively connected to a remote server to implement data transmission with the server. The server may include a server on the task scheduling center side, and in other implementation scenarios, the server may also include a server on an intermediate platform, for example, a server on a third-party server platform that is communicatively linked to the task scheduling center server. The server may include a single computer device, or may include a server cluster formed by a plurality of servers, or a server structure of a distributed apparatus.

The server and the client device may communicate using any suitable network protocol, including network protocols not yet developed at the filing date of this application. The network protocol may include, for example, a TCP/IP protocol, a UDP/IP protocol, an HTTP protocol, an HTTPS protocol, or the like. Of course, the network Protocol may also include, for example, an RPC Protocol (Remote Procedure Call Protocol), a REST Protocol (Representational State Transfer Protocol), and the like used above the above Protocol.

The following embodiments and application examples are specifically and individually described in detail.

In order to solve the problems of low accuracy, low intelligence degree, low efficiency and the like in the existing portfolio selection process, the present application provides an embodiment of a portfolio data processing method, which is executed by a portfolio data processing device and specifically includes the following contents, referring to fig. 2:

step 100: and screening among a plurality of preset objective functions and a plurality of constraint conditions according to the target investment requirement information to form an investment portfolio strategy selection model comprising the objective functions and at least one constraint condition.

In step 100, the objective function may include: the specific examples of the maximum profitability of the investment portfolio, the minimum investment cost, the minimum maximum withdrawal of the investment portfolio, the minimum risk factor of the investment portfolio less than a set value, etc. are as follows:

(1) and (3) maximizing the income of the investment portfolio: under the condition that the total capital scale of the investment portfolio is fixed, the total income of each investment item held by the investment portfolio is maximized, and the objective function is as follows:

Wherein F is the total profit amount, T is the total profit calculation days, K_iAmount of return, Δ t, for a single day of i investment holder_iIs daily.

(2) Investment cost minimization: under the condition that the total income of the investment portfolio is fixed, the total investment amount of the investment portfolio is minimized, and the objective function is as follows:

wherein E is the total investment, T is the total number of days, Q_iInvestment amount, Δ t, for i investment item holding a single day_iIs daily.

(3) Maximum withdrawal of portfolio minimization: earning in portfolioIn the calculation interval, the value of subtracting the minimum unit net value from the maximum unit net value of the investment portfolio is minimized, and the income fluctuation is reduced, wherein the objective function is as follows: u ═ min [ maxN_t-minN_t]Where U is maximum withdrawal, N_tIs a t day unit net value and has the calculation formula of

Wherein A is_tTotal scale of taken-up of the portfolio for t days, E_tThe total investment cost of the investment portfolio for t days.

In step 100, the constraint condition may include: specific examples of the total investment amount of the investment portfolio being less than the total amount of the collected fund, the position proportion of the single investment in the investment portfolio not exceeding the maximum position proportion (for example, 60 percent and the like), the profit time range between the interest date and the expiration date of the financial product and the non-unique type of the investment are as follows:

(1) The total investment amount of the investment portfolio is less than the total amount of the recruited funds: e is more than or equal to 0 and less than or equal to E_maxWherein E is the total investment amount, E_maxTo collect a total amount of funds.

(2) The position holding proportion of a single investment product does not exceed the maximum position holding proportion: h_i≤H_maxIn which H is_iIs the position holding proportion of the ith investment.

(3) The profit time ranges between the financial product's origination date and expiration date: t is_min≤T≤T_maxWherein T is_minFor the day of rest, T_maxIs due date.

Step 200: and taking the investment data of each target investment object as a training sample, training the investment portfolio strategy selection model based on a preset multi-target particle swarm algorithm, obtaining a plurality of target solutions of the investment portfolio strategy selection model for representing the investment portfolio strategy, and generating investment portfolio strategy recommendation result data containing each target solution.

In step 200, a multi-target Particle Swarm Algorithm (Particle Swarm Optimization Algorithm) can be applied to train according to the multi-target multi-constraint investment portfolio strategy selection model constructed in the last step, future prediction trend data of various open market investment products are input as training samples, and a plurality of better solutions, namely a plurality of better investment portfolio strategies, are obtained through training for traders to select.

As can be seen from the above description, the investment portfolio data processing method provided in the embodiment of the present application obtains the investment portfolio strategy recommendation result by using the investment portfolio strategy selection model including a plurality of objective functions and at least one constraint condition, so as to effectively improve the efficiency and the intelligence degree of the investment portfolio data processing process, effectively improve the accuracy and the reliability of the investment portfolio recommendation result, and solve the problems of the existing method, such as high dependence on human factors, poor selection result, and high fluctuation rate, so as to provide reliable and stable transaction decisions for users, such as traders, effectively improve the profitability of asset management services, reduce investment risks, and improve user experience.

In order to improve the reliability and the intelligence degree of the solution process of the investment portfolio strategy selection model, in an embodiment of the investment portfolio data processing method provided by the present application, referring to fig. 3, the steps 200 in the investment portfolio data processing method specifically include the following contents:

step 210: and generating an initial population corresponding to the multi-target particle swarm algorithm according to each target investment object.

Step 220: and initializing a preset external file according to the initial population.

Step 230: a speed position updating step: and updating the speed and the position of the particles in the external file to form a new particle population, and storing the new particle population into a preset temporary population.

In step 230, a speed location update operation may be performed on the population in the external profile, i.e., a better investment strategy solution is found, and the obtained new population is stored in the temporary population

Wherein the velocity update formula is:

v_i(k+1)＝ωv_i(k)+c₁r₁[pB_i-x_i(k)]+c₂r₂[gB-x_i(k)]

x_i(k+1)＝v_i(k+1)+x_i(k)

wherein i represents a particle index, k represents a time index, v_i(k) Velocity, x, of the ith particle in the k generation_i(k) Is the position of the ith particle in the kth generation, ω is the inertial weight, c₁And c₂Is an acceleration constant, r₁And r₂Is [0,1 ]]Random number within the interval, pB_iAnd gB is the historical optimal position of the particle swarm, namely the optimal investment strategy generated in the iterative process.

Step 240: and forming a mixed population according to the temporary population and the initialization population.

Step 250: and acquiring the current global optimal position and the current individual optimal position of the mixed population.

Step 260: and adding 1 to the current iteration number, judging whether the new iteration number is greater than a preset training cycle number, and if so, executing the step 210.

Step 270: and outputting the current global optimal position and the current individual optimal position of the mixed population to be used as the current target solution of the investment portfolio strategy selection model.

If the new iteration number is less than or equal to the training loop number, the step 240 is executed again: and updating the speed position.

As can be seen from the above description, the investment portfolio data processing method provided in the embodiment of the present application solves the investment portfolio strategy selection model by applying the multi-objective particle swarm algorithm, so as to effectively improve the reliability and the intelligence degree of the solution process of the investment portfolio strategy selection model, and further effectively improve the accuracy and the effectiveness of obtaining a plurality of target solutions of the investment portfolio strategy selection model for representing the investment portfolio strategy.

In order to further improve the applicability, reliability and intelligence of the portfolio data processing, in an embodiment of the portfolio data processing method provided in the present application, referring to fig. 4, step 210 of the portfolio data processing method specifically includes the following steps:

step 211: and setting a plurality of particles according to the target investment objects, wherein each particle is used for representing a unique investment scheme of the uniquely corresponding target investment object within a preset time range.

Step 212: and combining the particles to form a particle population which comprises all investment strategy combinations corresponding to the target investment objects.

It is understood that the unique investment schemes of the respective particles for representing the uniquely corresponding target investment objects within the preset time range are as follows: each particle can represent only one target investment object (e.g., investment) and each particle can represent only one investment scenario. It is understood that the preset time range can be set according to practical application situations, for example: 24 hours, 12 hours, or two days, etc. The investment scenario may refer to: the investment amount or the investment percentage, etc. may be specifically set according to the actual application, and this is not limited in this application.

The population of particles (which may also be referred to as a population of particles or a population) represents all possible investment strategy solutions for each of the target investment objects. That is, in this scenario, each particle represents an investment strategy, i.e., the daily investment amount per investment, and the population represents all possible solutions to the investment strategy. Setting population scale, wherein the maximum iteration number, constants (inertia constants, learning factors) in the velocity position updating formula, the size of the individual external files, the size of the global external files, the initialization population, the temporary population and the mixed population.

Step 213: and carrying out initialization processing on the particle population to obtain an initial population.

In step 213, the value of each particle may be initialized in a random manner to generate an initial population.

As can be seen from the above description, the investment portfolio data processing method provided in the embodiment of the present application can effectively improve the pertinence of the multi-target particle swarm algorithm application process by setting the plurality of particles according to each target investment object, can provide personalized customized service for the user who provides the investment portfolio policy recommendation request, improves the user experience, and further can further improve the applicability, reliability, and intelligent degree of the investment portfolio data processing.

In order to improve the reliability and effectiveness of initializing the preset external profile, referring to fig. 5, an embodiment of the portfolio data processing method provided herein, wherein step 220 of the portfolio data processing method specifically comprises the following steps:

step 221: and determining an objective function value of each particle in the initial population based on the investment portfolio strategy selection model, and determining a fitness function value of each particle in the initial population based on a preset particle swarm fitness function.

Step 222: and storing the particles with the fitness function value smaller than 0 into a preset external archive, selecting an individual optimal position and a global optimal position in the sequence from small fitness function values to large fitness function values of the particles in the external archive, and setting the current iteration number as 0.

Specifically, the objective function value of each particle in the initial population can be calculated, the fitness value of each particle is calculated according to the maximum and minimum fitness function in the second step, the particle with the fitness value smaller than 0 is stored in an external file as the fitness value smaller than 0 represents that the particle is a non-dominant solution, initialization of the external file is completed, the individual optimal position and the global optimal position are selected for the particles in the external file according to the sequence from small to large of the fitness function values, and the number of operation iterations is set to be 0.

As can be seen from the above description, in the investment portfolio data processing method provided in the embodiment of the present application, the objective function value of each particle in the initial population is determined based on the investment portfolio policy selection model, and the fitness function value of each particle in the initial population is determined based on the preset particle swarm fitness function, so that the reliability and effectiveness of initializing the preset external archive can be effectively improved, and the reliability and effectiveness of processing the investment portfolio data can be further improved.

In order to effectively improve the reliability and effectiveness of forming the mixed population, in an embodiment of the portfolio data processing method provided in the present application, referring to fig. 6, step 240 in the portfolio data processing method specifically includes the following contents:

step 241: and determining the fitness function value of each particle in the temporary population based on the particle swarm fitness function.

Step 242: and mixing the temporary population and the initialization population to form a mixed population.

Specifically, the method comprises the following steps: calculating the fitness function value of the particles in the temporary population, and mixing the temporary population and the initialized population to generate a mixed population

As can be seen from the above description, in the investment portfolio data processing method provided in the embodiment of the present application, the fitness function value of each particle in the temporary population is determined based on the particle swarm fitness function, so that the reliability and effectiveness of forming a mixed population can be effectively improved, and the reliability and effectiveness of processing investment portfolio data can be further improved.

In order to effectively improve the reliability and accuracy of obtaining the current global optimal position and the current individual optimal position of the mixed population, in an embodiment of the method for processing investment portfolio data provided by the present application, referring to fig. 7, step 250 in the method for processing investment portfolio data specifically includes the following steps:

Step 251: and performing tournament selection processing on the mixed population to generate a global external profile and an individual external profile corresponding to the mixed population.

Step 252: and selecting the current global optimal position of the mixed population from the global external files, and selecting the current individual optimal position of the mixed population from the individual external files.

Specifically, a tournament selection operation is performed on the mixed population, two particles are randomly selected each time to compare fitness function values, the better particles are added into the global external archive until the global external archive is filled, specifically, the fitness function values of the particles in the mixed population generated in the step 5 are judged, if the fitness function values are larger than or equal to 0, the particles are directly discarded, if the fitness function values are smaller than 0, the particles are directly added if the individual external archive is not full, and if the individual external archive is full, the particles with the size of the individual external archive are intercepted and reserved after the individual external archive is sorted from small to large according to the fitness function values. And selecting the individual optimal position and the global optimal position of the current population from the individual external file and the global external file, namely the particle with the minimum fitness function value.

As can be seen from the above description, the investment portfolio data processing method provided in the embodiment of the present application performs tournament selection processing on the mixed population to generate the global external profile and the individual external profile corresponding to the mixed population, so as to effectively improve the reliability and accuracy of obtaining the current global optimal position and the individual optimal position of the mixed population, and further improve the reliability and accuracy of investment portfolio data processing.

In order to improve the intelligence degree and the applicability of the recommendation of the portfolio strategy, in an embodiment of the portfolio data processing method provided in the present application, referring to fig. 8, the step 100 of the portfolio data processing method further comprises the following steps:

step 010: and receiving a portfolio strategy recommendation request, wherein the portfolio strategy recommendation request comprises target investment requirement information and a plurality of specified target investment objects.

Correspondingly, the method for processing investment portfolio data after step 200 further comprises the following steps:

step 300: and sending the investment portfolio strategy recommendation result data to the sender of the investment portfolio strategy recommendation request.

As can be seen from the above description, the investment portfolio data processing method provided in the embodiment of the present application can effectively improve the intelligence degree and the applicability of investment portfolio policy recommendation, improve the user experience, and further improve the intelligence degree and the applicability of investment portfolio data processing by receiving the investment portfolio policy recommendation request and sending the data of the investment portfolio policy recommendation result to the sender of the investment portfolio policy recommendation request.

In terms of software, in order to solve the problems of low accuracy, low intelligence, and low efficiency in the existing portfolio selection process, the present application provides an embodiment of a portfolio data processing apparatus for executing all or part of the content in the portfolio data processing method, which specifically includes the following content, with reference to fig. 9:

the model selection module 10 is configured to perform screening among a plurality of preset objective functions and a plurality of preset constraint conditions according to the target investment requirement information to form an investment portfolio strategy selection model including the plurality of objective functions and at least one constraint condition.

In the model selection module 10, the objective function may include: the maximum profitability of the investment portfolio, the minimum investment cost, the minimum maximum withdrawal of the investment portfolio, the risk factor of the investment portfolio less than the set value, etc., and the constraint conditions may include: the total investment amount of the investment portfolio is less than the total amount of the collected fund, the position holding proportion of the single investment product does not exceed the maximum position holding proportion (such as 60 percent and the like), the income time range is between the starting date and the due date of the financial product, the types of the investment products are not unique, and the like.

And the strategy recommendation module 20 is configured to use the investment data of each target investment object as a training sample, train the investment portfolio strategy selection model based on a preset multi-target particle swarm algorithm, obtain a plurality of target solutions of the investment portfolio strategy selection model, which are used for representing an investment portfolio strategy, and generate investment portfolio strategy recommendation result data including each target solution.

In the strategy recommendation module 20, a multi-target particle swarm algorithm can be applied to training according to the established multi-target multi-constraint investment portfolio strategy selection model, future prediction trend data of various open market investment products are input as training samples, and a plurality of better solutions, namely a plurality of better investment portfolio strategies, are obtained through training for traders to select.

The embodiment of the investment portfolio data processing apparatus provided in the present application may be specifically configured to execute the processing flow of the embodiment of the investment portfolio data processing method in the foregoing embodiment, and the functions thereof are not described herein again, and refer to the detailed description of the embodiment of the method described above.

As can be seen from the above description, the investment portfolio data processing apparatus provided in the embodiment of the present application obtains the investment portfolio strategy recommendation result by using the investment portfolio strategy selection model including a plurality of objective functions and at least one constraint condition, so as to effectively improve the efficiency and the intelligence degree of the investment portfolio data processing process, effectively improve the accuracy and the reliability of the investment portfolio recommendation result, and solve the problems of the existing method, such as high dependence on human factors, poor selection result, and high fluctuation rate, thereby providing reliable and stable transaction decisions for users, such as traders, effectively improving the profitability of asset management services, reducing the investment risk, and improving the user experience.

For further explanation of the scheme, the application also provides a specific application example of the investment portfolio data processing method, aiming at the defects existing in the investment portfolio selection in the field of the existing bank asset management business, and the problems of high dependence degree of human factors, poor selection result, high fluctuation rate and the like existing in the existing method are not solved.

Investment portfolio strategy selection model

A investment portfolio strategy selection model based on multi-target multi-constraint conditions of bank asset management business transaction elements comprises the following specific contents: selecting a plurality of proper objective functions and constraint conditions by combining the actual business transaction objectives of the investment portfolio, wherein the objective functions can comprise: the specific examples of the maximum profitability of the investment portfolio, the minimum investment cost, the minimum maximum withdrawal of the investment portfolio, the minimum risk factor of the investment portfolio less than a set value, etc. are as follows:

(4) And (3) maximizing the income of the investment portfolio: under the condition that the total capital scale of the investment portfolio is fixed, the total income of each investment item held by the investment portfolio is maximized, and the objective function is as follows:

(5) Investment cost minimization: under the condition that the total income of the investment portfolio is fixed, the total investment amount of the investment portfolio is minimized, and the objective function is as follows:

(6) Maximum withdrawal of portfolio minimization: in the investment portfolio profit calculation interval, the value of subtracting the minimum unit net value from the maximum unit net value of the investment portfolio is minimum, and the profit fluctuation is reduced, wherein the objective function is as follows: u ═ min [ maxN_t-minN_t]Where U is maximum withdrawal, N_tIs a t day unit net value and has the calculation formula of

The constraints may include: specific examples of the total investment amount of the investment portfolio being less than the total amount of the collected fund, the position proportion of the single investment in the investment portfolio not exceeding the maximum position proportion (for example, 60 percent and the like), the profit time range between the interest date and the expiration date of the financial product and the non-unique type of the investment are as follows:

(4) The total investment amount of the investment portfolio is less than the total amount of the recruited funds: e is more than or equal to 0 and less than or equal to E_maxWherein E is the total investment amount, E_maxTo collect a total amount of funds.

(5) The position holding proportion of a single investment product does not exceed the maximum position holding proportion: h_i≤H_maxIn which H is_iIs the position holding proportion of the ith investment.

(6) The profit time ranges between the financial product's origination date and expiration date: t is_min≤T≤T_maxWherein T is_minFor the day of rest, T_maxIs due date.

Investment combination data processing method

Based on the above, referring to fig. 10, an application example of the present application provides an investment portfolio data processing method based on an artificial intelligence Algorithm, which includes the steps of training by applying a multi-target Particle Swarm Optimization (Particle Swarm Optimization Algorithm) according to an investment portfolio strategy selection model of a multi-target multi-constraint condition constructed in the previous step, inputting future prediction trend data of various open market investment products as a training sample, and obtaining a plurality of better solutions through training, that is, a plurality of better investment portfolio strategies for traders to select, and specifically includes:

(1) and determining parameters of the multi-target particle swarm algorithm. In this scenario, each particle represents an investment strategy, i.e., the daily investment amount per investment, and the population represents all possible investment strategy solutions. The population size P is set, where the maximum number of iterations G, the constants in the velocity position update formula (inertia constant ω, learning factor c) ₁And c₂) Individual external file is largeSmall NG, size of Global external archive NP, initialization population pop_kTemporary population

And mixed populations

The population sizes were P, P, and 2P, respectively.

(2) And selecting a particle swarm Fitness function, wherein a maximum and minimum Fitness function (Maximin Fitness) is selected as the Fitness function of the multi-target particle swarm algorithm.

(3) Initializing a population, initializing the value of each particle by adopting a random mode, calculating the objective function value of each particle in the initial population, calculating the fitness value of each particle according to the maximum and minimum fitness function in the second step, storing the particles with the fitness value smaller than 0 into an external file because the fitness value smaller than 0 represents that the particles are non-dominant solutions, finishing the initialization of the external file, selecting the individual optimal position and the global optimal position for the particles in the external file according to the order from small to large of the fitness function values, and setting the running iteration number to be 0.

(4) Performing speed and position updating operation on the population in the external archive, namely searching a better investment strategy solution, and storing the obtained new population in a temporary population

Wherein the velocity update formula is:

v_i(k+1)＝ωv_i(k)+c₁r₁[pB_i-x_i(k)]+c₂r₂[gB-x_i(k)]

x_i(k+1)＝v_i(k+1)+x_i(k)

wherein i represents a particle index, k represents a time index, v _i(k) Velocity, x, of the ith particle in the k generation_i(k) Is the position of the ith particle in the kth generation, ω is the inertial weight, c₁And c₂Is an acceleration constant, r₁And r₂Is [0,1 ]]Random number within the interval, pB_iAnd gB is the historical optimal position of the particle swarm, namely the optimal investment strategy generated in the iterative process.

(5) Calculating the fitness function value of the particles in the temporary population, and mixing the temporary population and the initialized population to generate a mixed population

(6) Maintaining an external file: and (2) performing a tournament selection operation on the mixed population, randomly selecting two particles at each time to compare fitness function values, adding better particles into the global external archive until the global external archive is filled, specifically, judging the fitness function values of the particles in the mixed population generated in the step 5, directly discarding the particles if the fitness function values are more than or equal to 0, directly adding the particles if the individual external archive is not full, and intercepting the particles with the sizes of the individual external archive after sorting the particles from small to large according to the fitness function values for reservation if the individual external archive is full.

(7) And selecting the individual optimal position and the global optimal position of the current population from the individual external file and the global external file, namely the particle with the minimum fitness function value.

(8) And adding 1 to the cycle times, judging whether a termination condition is reached, namely the set training cycle times, if so, outputting a solution set, otherwise, jumping to the step 4 to continue the cycle execution. And finally, the obtained non-inferior solution set consisting of a plurality of particles is the recommended investment strategy, and the strategy data is displayed on a page for traders to make reference decisions.

Based on the technical scheme, the application example provides an investment portfolio data processing method based on a multi-target multi-constraint artificial intelligence algorithm, and generation of recommendation results of investment portfolio selection strategies is achieved by constructing a multi-target multi-constraint strategy optimization model and application of a multi-target particle swarm algorithm. The method reduces the dependency on the business level of traders, improves the yield of financial products in the limiting conditions, stably outputs the recommended investment portfolio strategy and reduces the possible risks in the trading process.

In terms of hardware, in order to solve the problems of low accuracy, low intelligence degree, low efficiency and the like in the existing investment portfolio selection process, the present application provides an embodiment of an electronic device for implementing all or part of the contents in the investment portfolio data processing method, and the electronic device specifically includes the following contents:

Fig. 11 is a schematic block diagram of a system configuration of an electronic device 9600 according to an embodiment of the present application. As shown in fig. 11, the electronic device 9600 can include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. Notably, this FIG. 11 is exemplary; other types of structures may also be used in addition to or in place of the structure to implement telecommunications or other functions.

In one embodiment, the portfolio data processing functionality may be integrated into a central processor. Wherein the central processor may be configured to control:

In step 100, the objective function may include: the maximum profitability of the investment portfolio, the minimum investment cost, the minimum maximum withdrawal of the investment portfolio, the risk factor of the investment portfolio less than the set value, etc., and the constraint conditions may include: the total investment amount of the investment portfolio is less than the total amount of the collected fund, the position holding proportion of the single investment product does not exceed the maximum position holding proportion (such as 60 percent and the like), the income time range is between the starting date and the due date of the financial product, the types of the investment products are not unique, and the like.

As can be seen from the above description, in the electronic device provided in the embodiment of the present application, the investment portfolio strategy selection model including a plurality of objective functions and at least one constraint condition is used to obtain the investment portfolio strategy recommendation result, so that the efficiency and the intelligence degree of the investment portfolio data processing process can be effectively improved, the accuracy and the reliability of the investment portfolio recommendation result can be effectively improved, the problems of high human factor dependency, poor selection result, high fluctuation rate and the like in the existing method can be solved, and further, reliable and stable trading decisions can be provided for users such as traders, the profitability of asset management services can be effectively improved, the investment risk can be reduced, and the user experience can be improved.

In another embodiment, the portfolio data processing apparatus may be configured separately from the central processor 9100, for example, the portfolio data processing apparatus may be configured as a chip connected to the central processor 9100, and the portfolio data processing function may be realized by the control of the central processor.

As shown in fig. 11, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is noted that the electronic device 9600 also does not necessarily include all of the components shown in fig. 11; in addition, the electronic device 9600 may further include components not shown in fig. 11, which may be referred to in the prior art.

As shown in fig. 11, a central processor 9100, sometimes referred to as a controller or operational control, can include a microprocessor or other processor device and/or logic device, which central processor 9100 receives input and controls the operation of the various components of the electronic device 9600.

The memory 9140 can be, for example, one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. The information relating to the failure may be stored, and a program for executing the information may be stored. And the central processing unit 9100 can execute the program stored in the memory 9140 to realize information storage or processing, or the like.

The input unit 9120 provides input to the central processor 9100. The input unit 9120 is, for example, a key or a touch input device. Power supply 9170 is used to provide power to electronic device 9600. The display 9160 is used for displaying display objects such as images and characters. The display may be, for example, an LCD display, but is not limited thereto.

The memory 9140 can be a solid state memory, e.g., Read Only Memory (ROM), Random Access Memory (RAM), a SIM card, or the like. There may also be a memory that holds information even when power is off, can be selectively erased, and is provided with more data, an example of which is sometimes called an EPROM or the like. The memory 9140 could also be some other type of device. Memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application/function storage portion 9142, the application/function storage portion 9142 being used for storing application programs and function programs or for executing a flow of operations of the electronic device 9600 by the central processor 9100.

The memory 9140 can also include a data store 9143, the data store 9143 being used to store data, such as contacts, digital data, pictures, sounds, and/or any other data used by an electronic device. The driver storage portion 9144 of the memory 9140 may include various drivers for the electronic device for communication functions and/or for performing other functions of the electronic device (e.g., messaging applications, contact book applications, etc.).

The communication module 9110 is a transmitter/receiver 9110 that transmits and receives signals via an antenna 9111. The communication module (transmitter/receiver) 9110 is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as in the case of a conventional mobile communication terminal.

Based on different communication technologies, a plurality of communication modules 9110, such as a cellular network module, a bluetooth module, and/or a wireless local area network module, may be provided in the same electronic device. The communication module (transmitter/receiver) 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and receive audio input from the microphone 9132, thereby implementing ordinary telecommunications functions. The audio processor 9130 may include any suitable buffers, decoders, amplifiers and so forth. In addition, the audio processor 9130 is also coupled to the central processor 9100, thereby enabling recording locally through the microphone 9132 and enabling locally stored sounds to be played through the speaker 9131.

Embodiments of the present application further provide a computer-readable storage medium capable of implementing all the steps in the portfolio data processing method in the above embodiments, where the computer-readable storage medium stores thereon a computer program, and when the computer program is executed by a processor, the computer program implements all the steps of the portfolio data processing method in the above embodiments, where the execution subject is a server or a client, for example, the processor implements the following steps when executing the computer program:

As can be seen from the above description, the computer-readable storage medium provided in the embodiment of the present application obtains the investment portfolio policy recommendation result by using the investment portfolio policy selection model including a plurality of objective functions and at least one constraint condition, so that the efficiency and the intelligence degree of the investment portfolio data processing process can be effectively improved, the accuracy and the reliability of the investment portfolio recommendation result can be effectively improved, the problems of high human factor dependency, poor selection result, high fluctuation rate and the like in the existing method can be solved, and further reliable and stable transaction decisions can be provided for users such as traders, the profitability of asset management services can be effectively improved, the investment risk can be reduced, and the user experience can be improved.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The principle and the implementation mode of the invention are explained by applying specific embodiments in the invention, and the description of the embodiments is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims

1. A method for processing portfolio data, comprising:

2. The method for processing investment portfolio data according to claim 1, wherein the training of the investment portfolio strategy selection model based on a preset multi-objective particle swarm algorithm with the respective investment data of each target investment object as a training sample to obtain a plurality of target solutions of the investment portfolio strategy selection model for representing investment portfolio strategy comprises:

initializing a preset external file according to the initial population;

3. The investment portfolio data processing method of claim 2, wherein said generating an initial population corresponding to a multi-target particle swarm algorithm according to each of said target investment objects comprises:

4. The portfolio data processing method of claim 3, wherein the initializing a preset external profile based on the initial population comprises:

5. The portfolio data processing method of claim 4, wherein the forming a mixed population based on the temporal population and the initialization population comprises:

6. The portfolio data processing method of claim 5, wherein the obtaining of the current global optimal location and the individual optimal location of the mixed population comprises:

7. The method according to any one of claims 1 to 6, further comprising, before the filtering among a plurality of predetermined objective functions and a plurality of predetermined constraints according to the target investment requirement information to form a portfolio strategy selection model including a plurality of objective functions and at least one constraint:

8. An investment portfolio data processing apparatus comprising:

9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the portfolio data processing method of any one of claims 1 through 7 when executing the computer program.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the portfolio data processing method of any one of claims 1 through 7.