A product recommendation method and device based on robust optimization, equipment and medium

Abstract

The application belongs to the technical field of financial data processing, and discloses a product recommendation method and device based on robust optimization, equipment and medium; the method comprises the following steps: obtaining investor preference information to form an investor preference vector P; mapping the investor preference vector P into a constraint on the future price or yield distribution F of an investment product; substituting the constraint into a parameter estimation model to obtain a parameter to be estimated; substituting the parameter into an investment probability model to obtain the investment probability of the investor in the investment product v j The investment probability is used to determine the recommended target product according to the investment probability. The target fund product corresponding to the investor preference is determined by using the investment probability, so that the target fund product recommended to the target user is accurately positioned according to the investment condition of the user and the fund product.

Description

Technical Field

[0001] This invention belongs to the field of financial data processing technology, and specifically relates to a product recommendation method, apparatus, equipment and medium based on robust optimization. Background Technology

[0002] A public offering of fund (PFA) is a mutual fund that raises capital from the general public through public channels and primarily invests in securities. PFAs are established by pooling public funds through mass media to invest in securities. These funds are subject to strict legal regulation and adhere to industry standards regarding information disclosure, profit distribution, and operational restrictions.

[0003] Taking publicly offered funds sold on a certain fund website as an example, when investors screen fund products, they generally first consider choosing those that are more equity-oriented or more bond-oriented. They typically rank fund products under the corresponding category according to return and risk indicators such as yield, drawdown, and volatility provided by the website or app, selecting the top-ranked products. They then combine this with factors such as the fund company's size, the fund manager's qualifications, and other evaluation indicators to determine the final investment product. With the explosive growth in the number of publicly offered fund products, it is becoming increasingly difficult for investors to select suitable funds. Both investors and fund sales institutions need a more effective method for product recommendation.

[0004] According to regulatory requirements, fund product sales institutions require investors to complete a risk assessment questionnaire before actual investment, answering several questions related to the investor's risk-return preferences. The results are used to determine the investor's level of risk aversion. Typically, investors are categorized into five types: conservative, moderate, balanced, active, and aggressive. Fund products are also categorized into these five types according to certain risk-return measurement methods. Investors can only invest in fund products that match their risk preferences. However, this approach is flawed because the risk assessment questions answered by investors are insufficient to accurately depict their true and complete risk preferences. The categorization of investors and investment products is also too crude, resulting in significant differences in investor preferences and product performance within the same category. Therefore, existing fund product recommendation systems lack the ability to recommend products based on accurate user investment conditions. When investors try to match their investment conditions with public fund products, they lack a systematic and scientific method, often resulting in poor matching accuracy and negatively impacting their investment experience. Summary of the Invention

[0005] The purpose of this invention is to provide a product recommendation method, apparatus, device, and medium based on robust optimization, so as to solve the technical problem of poor recommendation matching and positioning accuracy in existing fund product recommendation systems.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a product recommendation method based on robust optimization, comprising:

[0008] Obtain investor preference information and form an investor preference vector. ;

[0009] Investor preference vector Mapped to the distribution of future prices or returns of investment products Constraints;

[0010] Substituting the constraints and historical investor investment records into the model of the parameters to be estimated, we obtain the parameters to be estimated. , Representative products The corresponding parameters to be estimated;

[0011] For investment products The estimated parameters Substituting the new investor preference vector into the investment probability model, we can obtain the investment products for new investors. Investment probability ;

[0012] Based on investment probability Identify the target products to recommend.

[0013] A further improvement of this invention lies in: acquiring investor preference information and forming an investor preference vector. In the steps, investor preference vector ; Subscript p The total amount of investor preference information obtained.

[0014] A further improvement of the present invention is as follows:

[0015] Investor preference vector Mapped to the distribution of future prices or returns of investment products The steps for setting constraints specifically include:

[0016] The constraints are constraints on the moments of the future price or return distribution F of the investment product; let Represents the vector of investor preferences To constrain the distribution of future prices or returns of investment products by the k-th order moment, let This represents the range of the k-th order moments of the distribution under constraints.

[0017] A further improvement of the present invention is as follows:

[0018] For investment products Substituting the constraints and historical investor investment records into the parameter model to be estimated, the parameters are obtained. In the following steps, the model of the parameters to be estimated is:

[0019]

[0020] in, Indicating past investors Investment products Records Indicates investors Invested in products , Indicates investors I have never invested in any products. N represents the total number of past investors; Indicates investors Constraints on the moments of each order of the distribution F of future prices or returns of investment products; Indicates investors An uncertain set of preferences.

[0021] A further improvement of the present invention is that the uncertainty set used to reflect investor preferences is an elliptic uncertainty set.

[0022] A further improvement of this invention is that the expression for the uncertain set is:

[0023]

[0024] in, For investors who originally collected the data The investor preference vector express The List, Indicates the first l Measurement error of each feature; subscript The total amount of investor preference information obtained; Indicates distance, Represents a given positive definite matrix The defined norm.

[0025] A further improvement of the present invention is as follows:

[0026] For new investors Assuming her / his preference vector is And the corresponding constraints he / she imposes on the moments of the distribution F of future prices or returns of the investment product. Therefore, for any product whose expected future price or return distribution is known, its moments can be calculated, and by finding those that meet the investor's requirements, the range of product choices can be narrowed down.

[0027] For the initial selection of an investment product The previous steps have already estimated that and new investors And the constraints of each moment of the distribution F of future prices or returns of investment products. The probability of investment by new investors can be obtained using a logistic regression model. :

[0028]

[0029] in, product The probability of.

[0030] A further improvement of this invention lies in: based on investment probability The steps to determine the recommended target product include:

[0031] Based on investment probability Sort the products from largest to smallest to determine the probability of investment. The largest one or more products are the recommended target products.

[0032] Secondly, the present invention provides a product recommendation device based on robust optimization, comprising:

[0033] The data acquisition module is used to obtain investor preference information and form an investor preference vector. ;

[0034] The constraint module is used to convert the investor preference vector Mapped to the distribution of future prices or returns of investment products Constraints;

[0035] The calculation module is used to substitute the constraints and historical investor investment records into the parameter model to be estimated, and obtain the parameters. , Representative products The corresponding parameters to be estimated; for investment products The estimated parameters Substituting the new investor preference vector into the investment probability model, we can obtain the investor's investment products. Investment probability ;

[0036] The recommendation module is used to determine investment probabilities. Identify the target products to recommend.

[0037] Thirdly, the present invention provides an electronic device including a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the aforementioned robust optimization-based product recommendation method.

[0038] Fourthly, the present invention provides a computer-readable storage medium storing at least one instruction that, when executed by a processor, implements the aforementioned robust optimization-based product recommendation method.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] This invention provides a product recommendation method, apparatus, device, and medium based on robust optimization, including: acquiring investor preference information and forming an investor preference vector. ; Investor preference vector Mapped to the distribution of future prices or returns of investment products Constraints; substitute these constraints and historical investor investment records into the parameter model to be estimated to obtain the parameters. , Representative products The corresponding parameters to be estimated; for investment products The estimated parameters Substituting the new investor preference vector into the investment probability model, we can obtain the investor's investment products. Investment probability Based on investment probability The invention identifies target products for recommendation based on current user preference vectors and the distribution of expected future prices or returns for investment products. Requirements for each order of moments The method calculates the parameters to be estimated through the parameter model, calculates the investment probability by substituting the parameters into the logistic regression model, and uses the investment probability to determine the corresponding target fund product. This allows the target fund product recommended to the target user to accurately position the user's own investment conditions and fund product, which helps the target user improve the accuracy of positioning.

[0041] This invention helps investors narrow down their choices from thousands of investment options, recommending products that best suit their preferences. These preferences include risk-return requirements, product attributes such as investment scope, size, establishment date, and trading currency, as well as various distinguishing factors related to the product issuer and investment manager. Attached Figure Description

[0042] The accompanying drawings, which form part of this specification, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0043] Figure 1This is a flowchart illustrating a product recommendation method based on robust optimization according to the present invention.

[0044] Figure 2 This is a schematic diagram of the structure of a product recommendation device based on robust optimization according to the present invention;

[0045] Figure 3 This is a structural block diagram of an electronic device according to the present invention. Detailed Implementation

[0046] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0047] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this invention is for describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.

[0048] Example 1

[0049] Please see Figure 1 As shown, this invention provides a product recommendation method based on robust optimization, comprising:

[0050] Obtain investor preference information; investor preference information is represented by a p-dimensional preference vector. R represents all real numbers; It can be either a continuous or discrete value. The vector P contains anything that reflects an investor's requirements for an investment product, and its specific value can be measured in any way; in one specific implementation, It can be done through questionnaires, inferences from past information, and deductions from the overall performance of groups with certain similarities, etc. It can reflect investors' requirements for investment products, such as expected returns (return ratio, return fluctuation range), and which specific products they will not purchase. In one specific implementation, the investor preference vector P can also include other investor information, such as age, gender, wealth accumulation level, and portfolio holdings. In one specific implementation, the p-dimensional preference vector can be set and modified by the product recommendation agency.

[0051] Investor preference vectors are assessed using evaluation methods. This is mapped to constraints on the distribution of future prices or returns of an investment product. For example, when such constraints are constraints on the moments of the distribution of future prices or returns of an investment product, let... Represents the vector of investor preferences To constrain the distribution of future prices or returns of investment products by the k-th order moment, let This represents the range of the k-th order moments of the distribution under this constraint. In one specific implementation, the evaluation method can be assessed by completing a questionnaire.

[0052] Given that investor preferences themselves are unobservable, and in real life they can only be obtained through estimation, guessing, and other methods, a vector reflecting investor preferences can be created. There will inevitably be some errors, making it difficult to accurately determine the range of products investors can invest in. To address this issue, this invention proposes a product recommendation method based on robust optimization, which first analyzes the preference vector... The analysis is performed using the indicators in the preference vector. In the space, construct a set that considers the feature measurement error and includes all possible values ​​of the feature. This set is called the uncertainty set.

[0053] The specific method for establishing an uncertain set is as follows: given a positive definite matrix... ,make express Defined norm, Representing distance; then the uncertain set defined in elliptic form is:

[0054]

[0055] in, For investors who originally collected the data The investor preference vector express The List, Indicates the first l Measurement error of each feature; subscript The total amount of investor preference information obtained; Indicates distance, Represents a given positive definite matrix The defined norm.

[0056] The expected future price or return distribution of an externally given investment product After obtaining relevant information, one can then discuss with investors regarding... Requirements for each order of moments The comparison is performed to select products that meet all or some of the constraints. In one specific implementation, the matrices of F fall into the user's corresponding G. k In this process, you can choose the corresponding investment products and initially select K products.

[0057] Furthermore, for the K products selected using the above method (or the products classified into K types), if the investment records of N investors are known, a matrix can be used. This indicates past investment records, among which Indicates investors Invested in products , Indicates investors I have never invested in any products. .

[0058] For each investment product The probability of an investment for a given user can be obtained using a logistic regression model.

[0059] Estimated parameters and new investors And the constraints of each moment of the distribution F of future prices or returns of investment products. The probability of investment by new investors can be obtained using a logistic regression model. :

[0060]

[0061]

[0062] in, product The probability of.

[0063] Parameter estimation for logistic regression typically employs the likelihood maximization function. Considering the errors contained in the investor preference vector, this invention employs a robust optimization method to obtain... The specific method utilizes the aforementioned constructed uncertain set. By first in the uncertain set Minimize the likelihood function internally, and then find a way to maximize the likelihood function to estimate it. ,Right now:

[0064]

[0065] in, Indicating past investors Investment products Records Indicates investors Invested in products , Indicates investors I have never invested in any products. N represents the total number of past investors; Indicates investors Constraints on the moments of each order of the distribution F of future prices or returns of investment products; Indicates investors An uncertain set of preferences.

[0066] After the above-mentioned parameter model is established, for a new investor preference vector To obtain the distribution of expected future prices or returns for investment products. Requirements for each order of moments Then, the parameters to be estimated can be obtained by substituting them into the model. , parameters Substitute into the investment probability model to obtain its investment probability. This can be used as a basis to prioritize recommending investment probabilities to investors. The largest product. In one specific implementation, it can also be based on investment probability. The system sorts products and recommends the top-ranked ones; the number of recommendations can be preset.

[0067] The product recommendation method provided in this embodiment is based on the current user preference vector and the distribution of expected future prices or returns of investment products. Requirements for each order of moments The method calculates the parameters to be estimated through the parameter model, substitutes the parameters to be estimated into the logistic regression model to calculate the investment probability, and uses the investment probability to determine the corresponding target fund product. This allows the target fund product recommended to the target user to accurately position the user's own investment conditions and fund product, which helps to improve the target user's investment experience.

[0068] This invention enables investors to scientifically select and recommend investment products. Its direct result is to narrow the range of products that investors can choose from among a vast number of products. This range also takes into account the impact of errors in measuring investor preferences. Among the selected products, it can also prioritize and recommend products with a high probability of past transactions to investors based on their preferences and historical transaction information.

[0069] Example 2

[0070] Please see Figure 1 As shown, this invention provides a product recommendation method based on robust optimization, comprising:

[0071] S1. Obtain investor preference information and form an investor preference vector. ;

[0072] S2, Transform the investor preference vector Mapped to the distribution of future prices or returns of investment products Constraints;

[0073] S3. Substitute the constraints and historical investor investment records into the parameter model to be estimated to obtain the parameters. , Representative products The corresponding parameters to be estimated; for investment products The estimated parameters Substituting the new investor preference vector into the investment probability model, we can obtain the investor's investment products. Investment probability ;

[0074] S4. Based on investment probability Identify the target products to recommend.

[0075] In one specific implementation, investor preference information is obtained to form an investor preference vector. In the steps, investor preference vector ; p The total amount of investor preference information set.

[0076] In one specific implementation, the investor preference vector is... Mapped to the distribution of future prices or returns of investment products The steps for setting constraints specifically include:

[0077] The constraints are constraints on the moments of the future price or return distribution F of the investment product; let Represents the vector of investor preferences To constrain the distribution of future prices or returns of investment products by the k-th order moment, let This represents the range of the k-th order moments of the distribution under constraints.

[0078] In one specific implementation, for investment products Substituting the constraints and historical investor investment records into the parameter model to be estimated, the parameters are obtained. In the following steps, the model of the parameters to be estimated is:

[0079]

[0080] in, Indicating past investors Investment products Records Indicates investors Invested in products , Indicates investors I have never invested in any products. N represents past investors. The total number; This represents the constraints that investors place on the future price or return distribution F of the investment product at each order of moments; Indicates investors An uncertain set of preferences.

[0081] In one specific implementation, the expression for the uncertain set is:

[0082]

[0083] in, For investors who originally collected the data The investor preference vector express The List, Indicates the first l Measurement error of each feature; subscript The total amount of investor preference information obtained; Indicates distance, Represents a given positive definite matrix The defined norm.

[0084] In one specific implementation, for new investors Assuming her / his preference vector is And the corresponding constraints he / she imposes on the moments of the distribution F of future prices or returns of the investment product. Therefore, for any product whose expected future price or return distribution is known, its moments can be calculated, and by finding those that meet the investor's requirements, the range of product choices can be narrowed down.

[0085] For the initial selection of an investment product The previous steps have already estimated that and new investors And the constraints of each moment of the distribution F of future prices or returns of investment products. The probability of investment by new investors can be obtained using a logistic regression model. :

[0086]

[0087] in, product The probability of.

[0088] In one specific implementation, based on the investment probability The steps to determine the recommended target product include:

[0089] Based on investment probability Sort the products from largest to smallest to determine the probability of investment. The largest one or more products are the recommended target products.

[0090] Example 3

[0091] Please see Figure 2 As shown, the present invention provides a product recommendation device based on robust optimization, comprising:

[0092] The data acquisition module is used to obtain investor preference information and form an investor preference vector. ;

[0093] The constraint module is used to convert the investor preference vector Mapped to the distribution of future prices or returns of investment products Constraints;

[0094] The calculation module is used to substitute the constraints and historical investor investment records into the model of the parameters to be estimated, and obtain the parameters to be estimated. , Representative products The corresponding parameters to be estimated; for investment products The estimated parameters Substituting the new investor preference vector into the investment probability model, we can obtain the investor's investment products. Investment probability ;

[0095] The recommendation module is used to determine investment probabilities. Identify the target products to recommend.

[0096] In one specific implementation, investor preference information is obtained to form an investor preference vector. In the steps, investor preference vector ; p The total amount of investor preference information set.

[0097] In one specific implementation, the investor preference vector is... Mapped to the distribution of future prices or returns of investment products The steps for setting constraints specifically include:

[0098] The constraints are constraints on the moments of the future price or return distribution F of the investment product; let Represents the vector of investor preferences To constrain the distribution of future prices or returns of investment products by the k-th order moment, let This represents the range of the k-th order moments of the distribution under constraints.

[0099] In one specific implementation, for investment products Substituting the constraints and historical investor investment records into the parameter model to be estimated, the parameters are obtained. In the following steps, the model of the parameters to be estimated is:

[0100]

[0101] in, Indicating past investors Investment products Records Indicates investors Invested in products , Indicates investors I have never invested in any products. N represents the total number of past investors; Indicates investors Constraints on the moments of each order of the distribution F of future prices or returns of investment products; Indicates investors An uncertain set of preferences.

[0102] In one specific implementation, the expression for the uncertain set is:

[0103]

[0104] in, For investors who originally collected the data The investor preference vector express The List, Indicates the first l Measurement error of each feature; subscript The total amount of investor preference information obtained; Indicates distance, Represents a given positive definite matrix The defined norm.

[0105] In one specific implementation, for new investors Assuming her / his preference vector is And the corresponding constraints he / she imposes on the moments of the distribution F of future prices or returns of the investment product. Therefore, for any product whose expected future price or return distribution is known, its moments can be calculated, and by finding those that meet the investor's requirements, the range of product choices can be narrowed down.

[0106] For the initial selection of an investment product The previous steps have already estimated that and new investors And the constraints of each moment of the distribution F of future prices or returns of investment products. The probability of investment by new investors can be obtained using a logistic regression model. :

[0107]

[0108] in, product The probability of.

[0109] In one specific implementation, based on the investment probability The steps to determine the recommended target product include:

[0110] Based on investment probability Sort the products from largest to smallest to determine the probability of investment. The largest one or more products are the recommended target products.

[0111] Example 4

[0112] Please see Figure 3 As shown, the present invention also provides an electronic device 100 for implementing a product recommendation method based on robust optimization; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0113] The memory 101 can be used to store the computer program 103. The processor 102 implements the steps of the robust optimization-based product recommendation method described in Embodiment 1 or 2 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device 100 (such as audio data), etc. In addition, the memory 101 may include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other non-volatile solid-state storage device.

[0114] The at least one processor 102 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or any conventional processor. The processor 102 is the control center of the electronic device 100, connecting various parts of the electronic device 100 via various interfaces and lines.

[0115] The memory 101 in the electronic device 100 stores multiple instructions to implement a robust optimization-based product recommendation method, and the processor 102 can execute the multiple instructions to achieve the following:

[0116] Obtain investor preference information and form an investor preference vector. ;

[0117] Investor preference vector Mapped to the distribution of future prices or returns of investment products Constraints;

[0118] Substituting the constraints and historical investor investment records into the model of parameters to be estimated, we obtain the parameters. , Representative products The corresponding parameters to be estimated; for investment products The estimated parameters Substituting the new investor preference vector into the investment probability model, we can obtain the investor's investment products. Investment probability ;

[0119] Based on investment probability Identify the target products to recommend.

[0120] Example 5

[0121] If the modules / units integrated in the electronic device 100 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0122] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can 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, etc.) containing computer-usable program code.

[0123] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can 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, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0124] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0125] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A product recommendation method based on robust optimization, characterized in that, include: Obtain investor preference information and form an investor preference vector. ; Investor preference vector Mapped to the distribution of future prices or returns of investment products Constraints; Substituting the constraints and historical investor investment records into the model of parameters to be estimated, we obtain the parameters. , Representative products The corresponding parameters to be estimated; for investment products The estimated parameters Substituting the new investor preference vector into the investment probability model, we can obtain the investor's investment products. Investment probability ; Based on investment probability Identify the target products to recommend; Investor preference vector Mapped to the distribution of future prices or returns of investment products The steps for imposing constraints specifically include: the constraints being constraints on the moments of the future price or return distribution F of the investment product; letting... Represents the vector of investor preferences To constrain the distribution of future prices or returns of investment products by the k-th order moment, let This represents the range of the k-th order moments of the distribution under constraints; For investment products Substituting the constraints and historical investor investment records into the parameter model to be estimated, the parameters are obtained. The model for the parameters to be estimated is: in, Indicating past investors Investment products Records Indicates investors Invested in products , Indicates investors I have never invested in any products. N represents past investors. The total number; This represents the constraints that investors place on the future price or return distribution F of the investment product at each order of moments; Indicates investors An uncertain set of preferences; For investment products The estimated parameters Substituting the new investor preference vector into the investment probability model, we can obtain the investor's investment products. Investment probability In the initial selection of a particular product for investment, The estimated parameters and new investors And the constraints of each moment of the distribution F of future prices or returns of investment products. The probability of investment by new investors can be obtained using a logistic regression model. : in, product The probability of.

2. The product recommendation method based on robust optimization according to claim 1, characterized in that, Obtain investor preference information and form an investor preference vector. In the steps, investor preference vector ; Subscript p The total amount of investor preference information obtained.

3. The product recommendation method based on robust optimization according to claim 1, characterized in that, The uncertainty set is an elliptic uncertainty set.

4. The product recommendation method based on robust optimization according to claim 1, characterized in that, Based on investment probability The steps to determine the recommended target product include: Based on investment probability Sort the products from largest to smallest to determine the probability of investment. The largest one or more products are the recommended target products.

5. A product recommendation device based on robust optimization, characterized in that, include: The data acquisition module is used to obtain investor preference information and form an investor preference vector. ; The constraint module is used to convert the investor preference vector Mapped to the distribution of future prices or returns of investment products Constraints; The calculation module is used to substitute the constraints and historical investor investment records into the parameter model to be estimated, and obtain the parameters. , Representative products The corresponding parameters to be estimated; for investment products The estimated parameters Substituting the new investor preference vector into the investment probability model, we can obtain the investor's investment products. Investment probability ; The recommendation module is used to determine investment probabilities. Identify the target products to recommend; Investor preference vector Mapped to the distribution of future prices or returns of investment products The steps for imposing constraints specifically include: the constraints being constraints on the moments of the future price or return distribution F of the investment product; letting... Represents the vector of investor preferences To constrain the distribution of future prices or returns of investment products by the k-th order moment, let This represents the range of the k-th order moments of the distribution under constraints; For investment products Substituting the constraints and historical investor investment records into the parameter model to be estimated, the parameters are obtained. The model for the parameters to be estimated is: in, Indicating past investors Investment products Records Indicates investors Invested in products , Indicates investors I have never invested in any products. N represents past investors. The total number; This represents the constraints that investors place on the future price or return distribution F of the investment product at each order of moments; Indicates investors An uncertain set of preferences; For investment products The estimated parameters Substituting the new investor preference vector into the investment probability model, we can obtain the investor's investment products. Investment probability In the initial selection of a particular product for investment, The estimated parameters and new investors And the constraints of each moment of the distribution F of future prices or returns of investment products. The probability of investment by new investors can be obtained using a logistic regression model. : in, product The probability of.

6. An electronic device, characterized in that, It includes a processor and a memory, the processor being used to execute a computer program stored in the memory to implement a product recommendation method based on robust optimization as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one instruction, which, when executed by a processor, implements a robust optimization-based product recommendation method as described in any one of claims 1 to 4.

Images