1. A recommendation method for a fund manager, the method comprising:

acquiring a fund manager factor standard score set of historical adjacent months of a target recommended month;

screening effective factors according to the fund manager factor standard score set by adopting a relevance comparison method of grading ranking and average profitability ranking to obtain an effective factor set;

and acquiring a preset factor combination rule, and recommending the fund managers according to the effective factor set and the fund manager factor standard scoring set by adopting the preset factor combination rule to obtain a recommended fund manager set.

2. The fund manager recommendation method according to claim 1, wherein the step of obtaining a fund manager factor criteria score set for historical adjacent months of the target recommended month comprises:

acquiring an initial scoring set of fund manager factors to be standardized of historical adjacent months of the target recommended month;

acquiring a factor from all factors corresponding to the initial scoring set of the fund manager factors to be standardized as the factor to be standardized;

acquiring a factor initial score from the fund manager factor initial score set to be standardized according to the factor to be standardized to obtain a single factor initial score set to be standardized;

standardizing the single-factor initial score set to be standardized to obtain a single-factor standard score subset;

repeatedly executing the step of acquiring one factor from all factors corresponding to the initial scoring set of the fund manager factors to be standardized as the factor to be standardized until the acquisition of all factors corresponding to the initial scoring set of the fund manager factors to be standardized is completed;

using all of the sub-set of single-factor criterion scores as the set of fund manager factor criterion scores for historical contiguous months of the target recommended month.

3. The recommendation method of a fund manager according to claim 1, wherein the step of screening effective factors according to the fund manager factor standard score set by using a relevance comparison method of a grading ranking and an average profitability ranking to obtain an effective factor set comprises:

acquiring a factor from all factors corresponding to the fund manager factor standard score set as a factor to be analyzed;

acquiring factor standard scores from the fund manager factor standard score set according to the factors to be analyzed to obtain a single factor standard score set to be analyzed;

acquiring a fund manager grading rule corresponding to the factor to be analyzed, and according to the fund manager grading rule and the single factor standard score set to be analyzed, grading all fund managers corresponding to the fund manager factor standard score set according to the fund managers to obtain fund manager sets corresponding to the fund manager grades respectively;

according to each fund manager set, performing grading average yield calculation on historical adjacent months of the target recommended month for each fund manager grade to obtain grading average yields corresponding to each fund manager grade and to be ranked;

performing ranking calculation according to the grading average profitability of each fund manager to be ranked to obtain a grading average profitability ranking corresponding to each fund manager grading;

acquiring the grading ranking of each fund manager grade, and calculating the correlation coefficient of the grading ranking and the average profitability ranking according to each grading ranking and each grading average profitability ranking to obtain the grading correlation coefficient corresponding to each fund manager grade;

acquiring a preset correlation judgment threshold, and performing factor validity judgment according to the preset correlation judgment threshold and each grading correlation coefficient to obtain a factor validity judgment result corresponding to the factor to be analyzed;

repeatedly executing the step of acquiring one factor from all factors corresponding to the fund manager factor standard score set as a factor to be analyzed until the acquisition of all factors corresponding to the fund manager factor standard score set is completed;

and taking all factors with the effective factor validity judgment result as the effective factor set.

4. The method according to claim 1, wherein the step of recommending a fund manager according to the effective factor set and the fund manager factor standard score set by using the preset factor combination rule to obtain a recommended fund manager set comprises:

obtaining a verification sample set, a validity index set and a monotonicity index set according to the target recommended month;

obtaining a Shanghai depth 300 index list, and verifying each factor in the effective factor set according to the Shanghai depth 300 index list, the verification sample set, the validity index set and the monotonicity index set to obtain a verification factor set;

performing factor combination score calculation according to the fund manager factor standard score set and the verification factor set by adopting the preset factor combination rule to obtain factor combination scores corresponding to the fund managers respectively;

and recommending fund managers according to the factor combination scores to obtain the recommended fund manager set.

5. The recommendation method of a fund manager, according to claim 4, wherein the step of obtaining a set of validation samples, a set of validity indicators and a set of monotonicity indicators according to the target recommended month comprises:

acquiring the sample extraction quantity and a verification sample database;

and acquiring a plurality of verification samples from the verification sample database as the verification sample set by taking the historical adjacent month of the target recommended month as the starting time in a past history acquisition mode, wherein the number of the verification samples in the verification sample set is the same as the sample extraction number.

6. The recommendation method of a fund manager, according to claim 4, wherein before the step of obtaining the verification sample set, the validity index set and the monotonicity index set according to the target recommendation month, further comprising:

acquiring an initial score set of fund manager factors to be extracted, a factor library and a sample extraction proportion, wherein the initial score set of fund manager factors to be extracted is an initial score set of fund manager factors of historical adjacent months of months to be extracted;

obtaining a factor from the factor library as a factor of a sample to be extracted;

according to the factors of the samples to be extracted, carrying out reverse ordering on the factor initial scores in the fund manager factor initial score set to be extracted to obtain an ordered fund manager factor initial score set;

extracting factor initial scores from the sorted fund manager factor initial score sets by adopting the sample extraction proportion, and taking the fund managers corresponding to all the extracted factor initial scores as fund manager sets to be preprocessed;

obtaining fund data according to the fund manager set to be preprocessed and the months to be extracted to obtain a fund data set to be analyzed;

carrying out extreme value removal on the fund manager set to be preprocessed according to the fund data set to be analyzed by adopting a median value removing method to obtain a fund manager set of a sample to be extracted;

acquiring fund data from the fund data set to be analyzed according to the fund manager set of the sample to be extracted to obtain a fund data set to be extracted;

respectively carrying out market value ratio calculation on the fund manager of each sample to be extracted in the fund manager set of the sample to be extracted according to the fund data set to be extracted to obtain the fund market value ratio to be analyzed corresponding to each fund manager of each sample to be extracted;

respectively carrying out weighted summation on the monthly yield of the fund according to the market value ratio of the fund to be analyzed and the fund data set to be extracted aiming at each fund manager of the sample to be extracted to obtain the monthly yield comprehensive value of the fund manager corresponding to each fund manager of the sample to be extracted;

generating verification samples according to the monthly yield comprehensive values of the fund managers to obtain a verification sample subset of the months to be extracted corresponding to the factors of the samples to be extracted;

repeatedly executing the step of obtaining a factor from the factor library as the factor of the sample to be extracted until the obtaining of the factor in the factor library is completed;

storing each verification sample subset of the month to be extracted corresponding to each factor in the factor library in the verification sample database.

7. The recommendation method of the fund manager, wherein the step of verifying each factor in the set of valid factors separately according to the Hui depth 300 index list, the verification sample set, the set of validity indicators and the set of monotonicity indicators to obtain the set of verified factors comprises:

acquiring a factor from the effective factor set as a factor to be verified;

obtaining a verification sample from the verification sample set according to the factor to be verified to obtain a verification sample set to be processed;

acquiring a preset single validation rule of the validity index, and respectively performing validity validation on each validity index in the validity index set according to the preset single validation rule of the validity index, the Shanghai depth 300 index list and the to-be-processed validation sample set to obtain a validity index validation result set;

acquiring a preset comprehensive validation rule of the validity index, and performing comprehensive validation on validity according to the preset comprehensive validation rule of the validity index and the validity index validation result set to obtain a comprehensive validation result;

acquiring a preset monotonicity index single item verification rule, and respectively performing monotonicity verification on each monotonicity index in the monotonicity index set according to the preset monotonicity index single item verification rule, the Shanghai depth 300 index list and the to-be-processed verification sample set to obtain a monotonicity index verification result set;

acquiring a preset monotonicity index comprehensive verification rule, and performing monotonicity comprehensive verification according to the preset monotonicity index comprehensive verification rule and the monotonicity index verification result set to obtain a monotonicity comprehensive verification result;

when the validity comprehensive verification result and the monotonicity comprehensive verification result both pass, determining the factor to be verified as a verification passing factor;

repeatedly executing the step of acquiring a factor from the effective factor set as a factor to be verified until all the factors in the effective factor set are acquired;

and taking all the factors passing the verification as the verification factor set.

8. A recommendation device for a fund manager, the device comprising:

the data acquisition module is used for acquiring a fund manager factor standard score set of historical adjacent months of the target recommended month;

the effective factor set acquisition module is used for screening effective factors according to the fund manager factor standard scoring set by adopting a relevance comparison method of grading ranking and average profitability ranking to obtain an effective factor set;

and the recommendation fund manager set determining module is used for acquiring a preset factor combination rule, and recommending the fund managers according to the effective factor set and the fund manager factor standard scoring set by adopting the preset factor combination rule to obtain a recommendation fund manager set.

9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 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 steps of the method of any one of claims 1 to 7.

Background

At present, because fund managers are more, in order to facilitate users to select the fund managers, all large fund companies and related companies recommend the fund managers after rating the fund managers. The method for recommending the fund managers after rating is characterized in that indexes such as fluctuation rate withdrawal of fund historical performance managed by the fund managers are compared, investment performance is mainly analyzed, characteristics of the fund managers in other aspects are ignored, and the fund managers are difficult to accurately recommend.

Disclosure of Invention

The main purpose of the present application is to provide a fund manager recommendation method, device, equipment and storage medium, which aim to solve the technical problem that in the prior art, fund manager recommendation is difficult to accurately recommend a fund manager due to neglecting the characteristics of other aspects of the fund manager by comparing indexes such as the volatility of fund historical performance managed by the fund manager.

In order to achieve the above object, the present application provides a recommendation method for fund managers, the method comprising:

acquiring a fund manager factor standard score set of historical adjacent months of a target recommended month;

screening effective factors according to the fund manager factor standard score set by adopting a relevance comparison method of grading ranking and average profitability ranking to obtain an effective factor set;

and acquiring a preset factor combination rule, and recommending the fund managers according to the effective factor set and the fund manager factor standard scoring set by adopting the preset factor combination rule to obtain a recommended fund manager set.

Further, the step of obtaining a fund manager factor standard score set of historical adjacent months of the target recommended month includes:

acquiring an initial scoring set of fund manager factors to be standardized of historical adjacent months of the target recommended month;

acquiring a factor from all factors corresponding to the initial scoring set of the fund manager factors to be standardized as the factor to be standardized;

acquiring a factor initial score from the fund manager factor initial score set to be standardized according to the factor to be standardized to obtain a single factor initial score set to be standardized;

standardizing the single-factor initial score set to be standardized to obtain a single-factor standard score subset;

repeatedly executing the step of acquiring one factor from all factors corresponding to the initial scoring set of the fund manager factors to be standardized as the factor to be standardized until the acquisition of all factors corresponding to the initial scoring set of the fund manager factors to be standardized is completed;

using all of the sub-set of single-factor criterion scores as the set of fund manager factor criterion scores for historical contiguous months of the target recommended month.

Further, the step of screening effective factors according to the fund manager factor standard score set by adopting a correlation comparison method of the grading ranking and the average profitability ranking to obtain an effective factor set comprises the following steps:

acquiring a factor from all factors corresponding to the fund manager factor standard score set as a factor to be analyzed;

acquiring factor standard scores from the fund manager factor standard score set according to the factors to be analyzed to obtain a single factor standard score set to be analyzed;

acquiring a fund manager grading rule corresponding to the factor to be analyzed, and according to the fund manager grading rule and the single factor standard score set to be analyzed, grading all fund managers corresponding to the fund manager factor standard score set according to the fund managers to obtain fund manager sets corresponding to the fund manager grades respectively;

according to each fund manager set, performing grading average yield calculation on historical adjacent months of the target recommended month for each fund manager grade to obtain grading average yields corresponding to each fund manager grade and to be ranked;

performing ranking calculation according to the grading average profitability of each fund manager to be ranked to obtain a grading average profitability ranking corresponding to each fund manager grading;

acquiring the grading ranking of each fund manager grade, and calculating the correlation coefficient of the grading ranking and the average profitability ranking according to each grading ranking and each grading average profitability ranking to obtain the grading correlation coefficient corresponding to each fund manager grade;

acquiring a preset correlation judgment threshold, and performing factor validity judgment according to the preset correlation judgment threshold and each grading correlation coefficient to obtain a factor validity judgment result corresponding to the factor to be analyzed;

repeatedly executing the step of acquiring one factor from all factors corresponding to the fund manager factor standard score set as a factor to be analyzed until the acquisition of all factors corresponding to the fund manager factor standard score set is completed;

and taking all factors with the effective factor validity judgment result as the effective factor set.

Further, the step of recommending the fund manager according to the effective factor set and the fund manager factor standard score set by using the preset factor combination rule to obtain a recommended fund manager set includes:

obtaining a verification sample set, a validity index set and a monotonicity index set according to the target recommended month;

obtaining a Shanghai depth 300 index list, and verifying each factor in the effective factor set according to the Shanghai depth 300 index list, the verification sample set, the validity index set and the monotonicity index set to obtain a verification factor set;

performing factor combination score calculation according to the fund manager factor standard score set and the verification factor set by adopting the preset factor combination rule to obtain factor combination scores corresponding to the fund managers respectively;

and recommending fund managers according to the factor combination scores to obtain the recommended fund manager set.

Further, the step of obtaining a verification sample set, a validity index set, and a monotonicity index set according to the target recommended month includes:

acquiring the sample extraction quantity and a verification sample database;

and acquiring a plurality of verification samples from the verification sample database as the verification sample set by taking the historical adjacent month of the target recommended month as the starting time in a past history acquisition mode, wherein the number of the verification samples in the verification sample set is the same as the sample extraction number.

Further, before the step of obtaining the verification sample set, the validity index set, and the monotonicity index set according to the target recommended month, the method further includes:

acquiring an initial score set of fund manager factors to be extracted, a factor library and a sample extraction proportion, wherein the initial score set of fund manager factors to be extracted is an initial score set of fund manager factors of historical adjacent months of months to be extracted;

obtaining a factor from the factor library as a factor of a sample to be extracted;

according to the factors of the samples to be extracted, carrying out reverse ordering on the factor initial scores in the fund manager factor initial score set to be extracted to obtain an ordered fund manager factor initial score set;

extracting factor initial scores from the sorted fund manager factor initial score sets by adopting the sample extraction proportion, and taking the fund managers corresponding to all the extracted factor initial scores as fund manager sets to be preprocessed;

obtaining fund data according to the fund manager set to be preprocessed and the months to be extracted to obtain a fund data set to be analyzed;

carrying out extreme value removal on the fund manager set to be preprocessed according to the fund data set to be analyzed by adopting a median value removing method to obtain a fund manager set of a sample to be extracted;

acquiring fund data from the fund data set to be analyzed according to the fund manager set of the sample to be extracted to obtain a fund data set to be extracted;

respectively carrying out market value ratio calculation on the fund manager of each sample to be extracted in the fund manager set of the sample to be extracted according to the fund data set to be extracted to obtain the fund market value ratio to be analyzed corresponding to each fund manager of each sample to be extracted;

respectively carrying out weighted summation on the monthly yield of the fund according to the market value ratio of the fund to be analyzed and the fund data set to be extracted aiming at each fund manager of the sample to be extracted to obtain the monthly yield comprehensive value of the fund manager corresponding to each fund manager of the sample to be extracted;

generating verification samples according to the monthly yield comprehensive values of the fund managers to obtain a verification sample subset of the months to be extracted corresponding to the factors of the samples to be extracted;

repeatedly executing the step of obtaining a factor from the factor library as the factor of the sample to be extracted until the obtaining of the factor in the factor library is completed;

storing each verification sample subset of the month to be extracted corresponding to each factor in the factor library in the verification sample database.

Further, the step of verifying each factor in the valid factor set according to the Shanghai depth 300 index list, the verification sample set, the validity index set, and the monotonicity index set to obtain a verification factor set includes:

acquiring a factor from the effective factor set as a factor to be verified;

obtaining a verification sample from the verification sample set according to the factor to be verified to obtain a verification sample set to be processed;

acquiring a preset single validation rule of the validity index, and respectively performing validity validation on each validity index in the validity index set according to the preset single validation rule of the validity index, the Shanghai depth 300 index list and the to-be-processed validation sample set to obtain a validity index validation result set;

acquiring a preset comprehensive validation rule of the validity index, and performing comprehensive validation on validity according to the preset comprehensive validation rule of the validity index and the validity index validation result set to obtain a comprehensive validation result;

acquiring a preset monotonicity index single item verification rule, and respectively performing monotonicity verification on each monotonicity index in the monotonicity index set according to the preset monotonicity index single item verification rule, the Shanghai depth 300 index list and the to-be-processed verification sample set to obtain a monotonicity index verification result set;

acquiring a preset monotonicity index comprehensive verification rule, and performing monotonicity comprehensive verification according to the preset monotonicity index comprehensive verification rule and the monotonicity index verification result set to obtain a monotonicity comprehensive verification result;

when the validity comprehensive verification result and the monotonicity comprehensive verification result both pass, determining the factor to be verified as a verification passing factor;

repeatedly executing the step of acquiring a factor from the effective factor set as a factor to be verified until all the factors in the effective factor set are acquired;

and taking all the factors passing the verification as the verification factor set.

The application also provides a recommendation device of fund manager, the device includes:

the data acquisition module is used for acquiring a fund manager factor standard score set of historical adjacent months of the target recommended month;

the effective factor set acquisition module is used for screening effective factors according to the fund manager factor standard scoring set by adopting a relevance comparison method of grading ranking and average profitability ranking to obtain an effective factor set;

and the recommendation fund manager set determining module is used for acquiring a preset factor combination rule, and recommending the fund managers according to the effective factor set and the fund manager factor standard scoring set by adopting the preset factor combination rule to obtain a recommendation fund manager set.

The present application further proposes a computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the steps of any of the above methods when executing the computer program.

The present application also proposes a computer-readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method of any of the above.

According to the fund manager recommending method, the fund manager recommending device, the fund manager recommending equipment and the fund manager recommending storage medium, the fund manager factor standard score set of historical adjacent months of a target recommended month is obtained, effective factors are screened according to the fund manager factor standard score set by using a relevance comparison method of grading ranking and average profitability ranking, a preset factor combination rule is obtained, the preset factor combination rule is adopted, fund manager recommending is carried out according to the effective factor set and the fund manager factor standard score set, the recommended fund manager set is obtained, and fund manager recommending from multiple factor dimensions is achieved, so that investment performance and other aspects of the fund manager are fully considered, and the fund manager recommending accuracy is improved.

Drawings

FIG. 1 is a schematic flow chart illustrating a recommendation method for a fund manager according to an embodiment of the present application;

FIG. 2 is a block diagram illustrating a recommendation apparatus of a fund manager according to an embodiment of the present application;

fig. 3 is a block diagram illustrating a structure of a computer device according to an embodiment of the present application.

The implementation, functional features and advantages of the objectives of the present application will be further explained with reference to the accompanying drawings.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

Referring to fig. 1, an embodiment of the present application provides a recommendation method for a fund manager, where the method includes:

s1: acquiring a fund manager factor standard score set of historical adjacent months of a target recommended month;

s2: screening effective factors according to the fund manager factor standard score set by adopting a relevance comparison method of grading ranking and average profitability ranking to obtain an effective factor set;

s3: and acquiring a preset factor combination rule, and recommending the fund managers according to the effective factor set and the fund manager factor standard scoring set by adopting the preset factor combination rule to obtain a recommended fund manager set.

In the embodiment, the fund manager recommendation is performed from multiple factor dimensions by obtaining the fund manager factor standard score set of historical adjacent months of the target recommended month, performing effective factor screening according to the fund manager factor standard score set by using a correlation comparison method of a grading ranking and an average profitability ranking, obtaining the effective factor set, obtaining the preset factor combination rule, and performing the fund manager recommendation according to the effective factor set and the fund manager factor standard score set by using the preset factor combination rule, so that the fund manager recommendation is performed in full consideration of the investment performance and other aspects of the fund manager, and the recommendation accuracy of the fund manager is improved.

For S1, the fund manager factor standard score set may be obtained from a database, the fund manager factor standard score set may be obtained from a third-party application system, or the fund manager factor standard score set input by the user may be obtained.

The target recommended month is the month in which the fund manager is required to recommend.

The historical adjacent months of the target recommended month refer to the historical adjacent months of the target recommended month. For example, the target recommended month is 2021 year 2 month, and the historical adjacent month of the target recommended month is 2021 year 1 month, which is not specifically limited by this example.

The fund manager is the manager responsible for fund management.

The fund manager factor standard score set is a fund manager factor standard score set of historical adjacent months of the target recommended month. The fund manager factor standard score set is data obtained by standardizing the initial score of each factor of each fund manager. The fund manager factor criteria score set comprises: fund manager identification and factor standard scoring subsets, wherein each fund manager identification corresponds to one factor standard scoring subset. The subset of factor criteria scores includes: factor identification and factor standard score, wherein each factor identification corresponds to one factor standard score. The fund manager identifier may be an identifier that uniquely identifies a fund manager, such as a fund manager name, fund manager ID, and the like. The factor identification may be a factor name, a factor ID, etc. that uniquely identifies a factor. The factors are derived from a pool of factors. The fund manager factor criteria score set is data determined from the fund manager's profile data. The image data of the fund manager includes image data (i.e., scores) of the fund manager in various factors.

The fund manager has six main categories of investment concept, process, team, product, achievement and public opinion, and each category has one or more factors. For example, the investment philosophy includes 6 factors, and the factor identifiers of the 6 factors are: the plate rotation, transaction frequency, holding period, risk cognition, configuration and individual stock selection total 6 factors, which is not specifically limited by the examples herein. For another example, the flow includes 6 factors, and the factor identifiers of the 6 factors are: internal research support, external research support, value style, process stability, capability expandability, and investment cycle, which are not specifically limited by the examples herein.

For S2, performing the grading ranking calculation and the average profitability ranking calculation of the fund manager respectively according to the fund manager factor standard score set, then calculating the correlation between the grading ranking and the average profitability ranking, obtaining a preset correlation judgment threshold, determining that the factor validity judgment result is valid when the correlation between the grading ranking and the average profitability ranking is smaller than the preset correlation judgment threshold, and determining that the factor validity judgment result is invalid when the correlation between the grading ranking and the average profitability ranking is larger than or equal to the preset correlation judgment threshold; and taking all factors with effective factor effectiveness judgment results as the effective factor set.

For S3, the preset factor combination rule may be obtained from the database, or the preset factor combination rule input by the user may be obtained, or the preset factor combination rule may be obtained from the third-party application system, or the preset factor combination rule may be written in the program implementing the present application.

Performing factor combination score calculation according to the effective factor set and the fund manager factor standard score set by adopting the preset factor combination rule to obtain factor combination scores corresponding to fund managers respectively; sorting all the factor combination scores in a descending order to obtain factor combination scores sorted in the descending order; acquiring a preset recommendation ratio, and extracting from the beginning of the descending-ordered factor combination scores by adopting the preset recommendation ratio to obtain a target factor combination score set; and taking the fund manager corresponding to each factor combination score in the target factor combination score set as the recommended fund manager set.

The preset factor combination rule includes but is not limited to: the average value is calculated.

For example, the preset recommendation proportion is 10%, and all fund managers corresponding to the factor combination scores of the top 10% of the factor combination scores after the descending sorting are used as a recommendation fund manager set, which is not specifically limited in this example.

In an embodiment, the step of obtaining the fund manager factor standard score set of historical adjacent months of the target recommended month includes:

s11: acquiring an initial scoring set of fund manager factors to be standardized of historical adjacent months of the target recommended month;

s12: acquiring a factor from all factors corresponding to the initial scoring set of the fund manager factors to be standardized as the factor to be standardized;

s13: acquiring a factor initial score from the fund manager factor initial score set to be standardized according to the factor to be standardized to obtain a single factor initial score set to be standardized;

s14: standardizing the single-factor initial score set to be standardized to obtain a single-factor standard score subset;

s15: repeatedly executing the step of acquiring one factor from all factors corresponding to the initial scoring set of the fund manager factors to be standardized as the factor to be standardized until the acquisition of all factors corresponding to the initial scoring set of the fund manager factors to be standardized is completed;

s16: using all of the sub-set of single-factor criterion scores as the set of fund manager factor criterion scores for historical contiguous months of the target recommended month.

According to the fund manager factor standard score set, the fund manager factor initial score set to be standardized of the historical adjacent months of the target recommended month is standardized and then serves as the fund manager factor standard score set of the historical adjacent months of the target recommended month, so that the scores in the fund manager factor standard score set are standardized, and the recommendation accuracy of the fund manager is further improved.

For S11, the initial score set of fund manager factors to be normalized of the historical adjacent month of the target recommended month may be obtained from the database, the initial score set of fund manager factors to be normalized of the historical adjacent month of the target recommended month input by the user may also be obtained, and the initial score set of fund manager factors to be normalized of the historical adjacent month of the target recommended month may also be obtained from a third party application system.

And S12, sequentially acquiring a factor from all factors corresponding to the initial scoring set of the fund manager factors to be standardized, and taking the acquired factor as the factor to be standardized.

For step S13, all factor initial scores corresponding to the factors to be normalized are obtained from the fund manager factor initial score set to be normalized, and the obtained all factor initial scores are used as the single factor initial score set to be normalized. That is, the factor initial scores in the single-factor initial score set to be normalized are all the factor initial scores corresponding to the factors to be normalized.

For S14, each factor initial score in the set of single factor initial scores to be normalized is normalized, and all the normalized scores are taken as the subset of the single factor standard scores. That is, each of the one-factor standard scores in the subset of the one-factor standard scores is data in an interval of 0 to 1, and each of the one-factor standard scores in the subset of the one-factor standard scores may be 0 or 1.

Optionally, a standard normal distribution method is adopted to normalize each factor initial score in the single factor initial score set to be normalized, and all normalized scores are used as the single factor standard score subset. All scores after normalization are normally distributed as the factor standard scores in the single-factor standard score subset.

For S15, repeating steps S12 to S15 until all factors corresponding to the initial score set of fund manager factors to be normalized are obtained.

For S16, all of the sub-sets of single factor criterion scores are taken as a set, which is taken as the set of fund manager factor criterion scores for historical contiguous months of the target recommended month.

In an embodiment, the step of screening effective factors according to the fund manager factor standard score set by using the method for comparing the relevance between the grading ranking and the average profitability ranking to obtain the effective factor set includes:

s21: acquiring a factor from all factors corresponding to the fund manager factor standard score set as a factor to be analyzed;

s22: acquiring factor standard scores from the fund manager factor standard score set according to the factors to be analyzed to obtain a single factor standard score set to be analyzed;

s23: acquiring a fund manager grading rule corresponding to the factor to be analyzed, and according to the fund manager grading rule and the single factor standard score set to be analyzed, grading all fund managers corresponding to the fund manager factor standard score set according to the fund managers to obtain fund manager sets corresponding to the fund manager grades respectively;

s24: according to each fund manager set, performing grading average yield calculation on historical adjacent months of the target recommended month for each fund manager grade to obtain grading average yields corresponding to each fund manager grade and to be ranked;

s25: performing ranking calculation according to the grading average profitability of each fund manager to be ranked to obtain a grading average profitability ranking corresponding to each fund manager grading;

s26: acquiring the grading ranking of each fund manager grade, and calculating the correlation coefficient of the grading ranking and the average profitability ranking according to each grading ranking and each grading average profitability ranking to obtain the grading correlation coefficient corresponding to each fund manager grade;

s27: acquiring a preset correlation judgment threshold, and performing factor validity judgment according to the preset correlation judgment threshold and each grading correlation coefficient to obtain a factor validity judgment result corresponding to the factor to be analyzed;

s28: repeatedly executing the step of acquiring one factor from all factors corresponding to the fund manager factor standard score set as a factor to be analyzed until the acquisition of all factors corresponding to the fund manager factor standard score set is completed;

s29: and taking all factors with the effective factor validity judgment result as the effective factor set.

In this embodiment, a method for comparing the relevance of the ranking and the average profitability ranking is adopted, and effective factor screening is performed according to the fund manager factor standard score set, so that factors that the relevance of the ranking and the average profitability ranking does not meet requirements are eliminated, and the factors that the relevance of the ranking and the average profitability ranking meets requirements are used as the effective factor set, thereby further improving the recommendation accuracy of the fund manager.

And S21, acquiring a factor from all factors corresponding to the fund manager factor standard score set, and taking the acquired factor as a factor to be analyzed.

And S22, acquiring all factor standard scores corresponding to the factors to be analyzed from the fund manager factor standard score set, and taking all the acquired factor standard scores as a single factor standard score set to be analyzed.

For S23, the fund manager grading rule corresponding to the factor to be analyzed may be obtained from a database, the fund manager grading rule corresponding to the factor to be analyzed may also be obtained from a third-party application system, the fund manager grading rule corresponding to the factor to be analyzed input by the user may also be obtained, and the fund manager grading rule corresponding to the factor to be analyzed may also be written in a program implementing the present application.

The step of dividing all fund managers corresponding to the fund manager factor standard score set according to the fund manager grading rule and the single factor standard score set to be analyzed according to the fund manager grading, so as to obtain fund manager sets corresponding to the fund manager grading respectively, namely dividing fund managers corresponding to the fund manager factor standard score set belonging to the same fund manager grading into one fund manager set according to the fund manager grading rule and the single factor standard score set to be analyzed, namely dividing the factor standard scores corresponding to the fund managers belonging to the same fund manager grading into the standardized score ranges of the same fund manager grading. Each fund manager is classified to correspond to a fund manager set.

For example, the fund manager adopts the grading rule: 5 fund managers are classified, and the standardized scoring ranges of the classification of the 5 fund managers are respectively as follows: (0, 0.2) (including 0, not including 0.2), [0.2,0.4) (including 0.2, not including 0.4), [0.4,0.6) (including 0.4, not including 0.6), [0.6,0.8) (including 0.6, not including 0.8), [0.8,1] (including 0.8, including 1), all fund managers corresponding to the fund manager factor standard rating set are classified by fund manager class, and the fund manager set corresponding to the fund manager class [0,0.2 ], the fund manager set corresponding to the fund manager class [0.2,0.4 ], the fund manager set corresponding to the fund manager class [0.4,0.6 ], the fund manager set corresponding to the fund manager class [0.6,0.8 ], the fund set corresponding to the fund manager class [0.6,0.8 ], and the fund manager set corresponding to the fund manager class [0.8, are not specifically exemplified.

For S24, acquiring one fund manager set from each fund manager set as a fund manager set to be calculated; acquiring a fund manager from the fund manager set to be calculated as a fund manager to be calculated; all funds managed by the fund manager to be calculated in historical adjacent months of the target recommended month are obtained from a fund library, and a fund set to be calculated is obtained; respectively carrying out market value ratio calculation on each fund to be calculated in the fund set to be calculated to obtain the market value ratio to be calculated corresponding to each fund to be calculated in the fund set to be calculated; weighting and summing monthly profitability of historical adjacent months of the target recommended month corresponding to each fund to be calculated in the fund set to be calculated by adopting the market value occupation ratio of each fund to be calculated to obtain monthly average profitability of a fund manager corresponding to the fund manager to be calculated; repeatedly executing the step of acquiring one fund manager from the fund manager set to be calculated as the fund manager to be calculated until the fund manager in the fund manager set to be calculated is acquired; carrying out average calculation on monthly average profitability of each fund manager to obtain the grading average profitability to be ranked corresponding to the fund manager grading corresponding to the fund manager set to be calculated; and repeating the step of acquiring one fund manager set from each fund manager set as the fund manager set to be calculated until the fund manager sets of each fund manager set are acquired.

And respectively calculating the market value ratio of each fund to be calculated in the fund set to be calculated to obtain the market value ratio of the fund to be calculated corresponding to each fund to be calculated in the fund set to be calculated, namely, dividing the market value of the target fund to be calculated by the total market value of the fund set to be calculated to obtain the market value ratio of the fund to be calculated corresponding to the target fund to be calculated, wherein the target fund to be calculated is any fund to be calculated in the fund set to be calculated.

Wherein, the step of weighting and summing monthly profitability of the historical adjacent months of the target recommended month corresponding to each fund to be calculated in the fund set to be calculated by adopting each fund to be calculated occupation ratio to obtain monthly average profitability of the fund manager corresponding to the fund to be calculated, for example, the fund set to be calculated has three funds J1, J2 and J3, the fund value occupation ratio to be calculated of J1 is Z1, the fund value occupation ratio to be calculated of J2 is Z2, the fund value occupation ratio to be calculated of J3 is Z3, the fund value of J1 is S1, the fund value of J2 is S8, the fund value of J3 is S3, and then (Z1 + Z2 × S3984 + Z3) is taken as the calculated monthly average profitability of the fund to be calculated by using the result of J1 + Z4642).

And S25, respectively carrying out ranking calculation on each grading average profitability to be ranked in all the grading average profitability to be ranked to obtain the grading average profitability ranking corresponding to each fund manager grading.

For S26, the ranking of each fund manager is that the ranking of the fund manager is 1 when the fund manager is ranked as the first rank, the ranking of the fund manager is 2 when the fund manager is ranked as the second rank, the ranking of the fund manager is 3 when the fund manager is ranked as the third rank, the ranking of the fund manager is 4 when the fund manager is ranked as the fourth rank, and the ranking of the fund manager is 5 when the fund manager is ranked as the fifth rank.

Wherein, the calculation formula of the grading correlation coefficient of the ith fund manager grading is F_i：

Wherein n is the total number of fund manager grades, x_iIs the rank ranking, y, of the ith fund manager's rank_iIs the ith said ranked average profitability ranking.

For S27, acquiring a preset starting comparison grading number and a preset ending comparison grading number; comparing the grading quantity with the grading ranking of each fund manager according to the preset starting grade quantity, and acquiring the grading correlation coefficient from each grading correlation coefficient to obtain a first grading correlation coefficient set; comparing the grading quantity with the grading ranking of each fund manager according to the preset ending, and obtaining the grading correlation coefficient from each grading correlation coefficient to obtain a second grading correlation coefficient set; and when each grading correlation coefficient in the first grading correlation coefficient set and each grading correlation coefficient in the second grading correlation coefficient set are smaller than the preset correlation judgment threshold value, determining that a factor validity judgment result corresponding to the factor to be analyzed is valid, otherwise, determining that the factor validity judgment result corresponding to the factor to be analyzed is invalid.

And the sum result of the preset starting comparison grading number and the preset ending comparison grading number is less than or equal to the grading number of the fund manager.

For example, the ranking ranks of the fund managers are 5, the preset starting comparison ranking number is 2, and the preset ending comparison ranking number is 2, the ranking correlation coefficient corresponding to the ranking name of 1 and the ranking correlation coefficient corresponding to the ranking name of 2 are used as a first ranking correlation coefficient set, and the ranking correlation coefficient corresponding to the ranking name of 4 and the ranking correlation coefficient corresponding to the ranking name of 5 are used as a second ranking correlation coefficient set, which is not specifically limited in this example.

For S28, steps S21 to S28 are repeatedly executed until the acquisition of all factors corresponding to the fund manager factor standard score set is completed.

At S29, all the factors whose result of the factor validity determination is valid are set as a set, and the set is set as the valid factor set.

In an embodiment, the step of performing fund manager recommendation according to the effective factor set and the fund manager factor standard score set by using the preset factor combination rule to obtain a recommended fund manager set includes:

s31: obtaining a verification sample set, a validity index set and a monotonicity index set according to the target recommended month;

s32: obtaining a Shanghai depth 300 index list, and verifying each factor in the effective factor set according to the Shanghai depth 300 index list, the verification sample set, the validity index set and the monotonicity index set to obtain a verification factor set;

s33: performing factor combination score calculation according to the fund manager factor standard score set and the verification factor set by adopting the preset factor combination rule to obtain factor combination scores corresponding to the fund managers respectively;

s34: and recommending fund managers according to the factor combination scores to obtain the recommended fund manager set.

According to the method and the device, the factors in the effective factor set are verified firstly, the preset factor combination rule is adopted, and factor combination score calculation and fund manager recommendation are sequentially carried out according to all verified factors, so that the effectiveness of the factors for the factor combination score calculation is further improved, and the recommendation accuracy of the fund manager is further improved.

For S31, the data in the verification sample set is a verification sample of history obtained from a verification sample database starting from a month share adjacent to the history of the target recommended month.

Verifying the sample database includes: factor identification, year and month, validation sample.

Validation samples include, but are not limited to: factor identification, fund manager identification, year and month, and fund manager monthly yield comprehensive value.

The validity index set and the monotonicity index set can be obtained from a database, the validity index set and the monotonicity index set can be obtained from a third-party application system, and the validity index set and the monotonicity index set input by a user can be obtained.

For S32, the hunch 300 index list may be obtained from a database, the hunch 300 index list may be obtained from a third-party application system, or the hunch 300 index list input by the user may be obtained.

The Shanghai depth 300 index List includes: year, month and Shanghai depth 300 index.

And respectively carrying out validity verification and monotonicity verification on each factor in the valid factor set according to the Shanghai depth 300 index list, the verification sample set, the validity index set and the monotonicity index set, wherein only the factor which passes the validity verification and the monotonicity verification at the same time is the factor which passes the verification, and all the factors which pass the verification are taken as the verification factor set.

And S33, performing factor combination score calculation according to the verification factor set and the fund manager factor standard score set by adopting the preset factor combination rule to obtain the factor combination scores corresponding to the fund managers respectively corresponding to the fund manager factor standard score set.

Optionally, one fund manager is obtained from all fund managers corresponding to the fund manager factor standard scoring set and serves as a fund manager to be combined for scoring; according to the fund manager to be combined and scored, acquiring the factor standard score from the fund manager factor standard score set to obtain a factor standard score set to be combined and scored; adopting the preset factor combination rule, and carrying out factor combination score calculation according to the factor standard score set and the verification factor set to be combined scored to obtain the factor combination score corresponding to the fund manager to be combined scored; and repeatedly executing the step of acquiring one fund manager as a fund manager to be combined and graded from all fund managers corresponding to the fund manager factor standard grading set until the extraction of all fund managers corresponding to the fund manager factor standard grading set is determined to be completed.

For S34, performing descending sorting on all the factor combination scores to obtain the factor combination scores after the descending sorting; extracting from the beginning of the descending sorted factor combination scores by adopting the preset recommendation proportion to obtain a target factor combination score set; and taking the fund manager corresponding to each factor combination score in the target factor combination score set as the recommended fund manager set.

In an embodiment, the step of obtaining the verification sample set, the validity index set, and the monotonicity index set according to the target recommended month includes:

s3111: acquiring the sample extraction quantity and a verification sample database;

s3112: and acquiring a plurality of verification samples from the verification sample database as the verification sample set by taking the historical adjacent month of the target recommended month as the starting time in a past history acquisition mode, wherein the number of the verification samples in the verification sample set is the same as the sample extraction number.

In this embodiment, a previous history obtaining manner is adopted, a plurality of verification samples are obtained from the verification sample database as the verification sample set, with the history adjacent month of the target recommended month as the start time, so as to provide a data basis for performing validity verification and monotonicity verification subsequently.

For S3111, the sample extraction number may be obtained from a database, may also be obtained from a third-party application system, may also be obtained from a sample extraction number input by a user, and may also be written in a program implementing the present application. The number of sample extractions is a specific number.

Optionally, the number of sample extractions is set to 72.

The verification sample database can be obtained from the database, can also be obtained from a third-party application system, and can also be obtained from the user input.

For S3112, a plurality of verification samples are obtained from the verification sample database in a manner of history acquisition using history adjacent months of the target recommended month as start times, and all the obtained verification samples are used as the verification sample set.

In an embodiment, before the step of obtaining the verification sample set, the validity index set, and the monotonicity index set according to the target recommended month, the method further includes:

s3121: acquiring an initial score set of fund manager factors to be extracted, a factor library and a sample extraction proportion, wherein the initial score set of fund manager factors to be extracted is an initial score set of fund manager factors of historical adjacent months of months to be extracted;

s3122: obtaining a factor from the factor library as a factor of a sample to be extracted;

s3123: according to the factors of the samples to be extracted, carrying out reverse ordering on the factor initial scores in the fund manager factor initial score set to be extracted to obtain an ordered fund manager factor initial score set;

s3124: extracting factor initial scores from the sorted fund manager factor initial score sets by adopting the sample extraction proportion, and taking the fund managers corresponding to all the extracted factor initial scores as fund manager sets to be preprocessed;

s3125: obtaining fund data according to the fund manager set to be preprocessed and the months to be extracted to obtain a fund data set to be analyzed;

s3126: carrying out extreme value removal on the fund manager set to be preprocessed according to the fund data set to be analyzed by adopting a median value removing method to obtain a fund manager set of a sample to be extracted;

s3127: acquiring fund data from the fund data set to be analyzed according to the fund manager set of the sample to be extracted to obtain a fund data set to be extracted;

s3128: respectively carrying out market value ratio calculation on the fund manager of each sample to be extracted in the fund manager set of the sample to be extracted according to the fund data set to be extracted to obtain the fund market value ratio to be analyzed corresponding to each fund manager of each sample to be extracted;

s3129: respectively carrying out weighted summation on the monthly yield of the fund according to the market value ratio of the fund to be analyzed and the fund data set to be extracted aiming at each fund manager of the sample to be extracted to obtain the monthly yield comprehensive value of the fund manager corresponding to each fund manager of the sample to be extracted;

s31210: generating verification samples according to the monthly yield comprehensive values of the fund managers to obtain a verification sample subset of the months to be extracted corresponding to the factors of the samples to be extracted;

s31211: repeatedly executing the step of obtaining a factor from the factor library as the factor of the sample to be extracted until the obtaining of the factor in the factor library is completed;

s31212: storing each verification sample subset of the month to be extracted corresponding to each factor in the factor library in the verification sample database.

In this embodiment, before the step of obtaining the verification sample set, the validity index set, and the monotonicity index set according to the target recommended month, each verification sample subset is generated month by month and stored in the verification sample database.

For S3121, the fund manager factor initial score set, the factor library, and the sample extraction ratio to be extracted may be obtained from a database, the fund manager factor initial score set, the factor library, and the sample extraction ratio to be extracted may also be obtained from a third-party application system, and the fund manager factor initial score set, the factor library, and the sample extraction ratio to be extracted, which are input by a user, may also be obtained.

The sample extraction ratio is a value of 0 to 1, and does not include 0 nor 1.

For S3122, a factor is obtained from the factor library, and the obtained factor is used as a factor of the sample to be extracted.

And for S3123, performing reverse ordering on the initial score set of the fund manager factors to be extracted according to the initial scores of the factors corresponding to the factors of the samples to be extracted to obtain an ordered initial score set of the fund manager factors. That is to say, the sorted fund manager factor initial score set is sorted in reverse order by the factor initial scores corresponding to the factors of the sample to be extracted.

For S3124, extracting the factor initial scores from the sorted fund manager factor initial score set by using the sample extraction ratio and a manner of starting extraction from the beginning, and taking the fund manager corresponding to each extracted factor initial score as the fund manager set to be preprocessed, that is, dividing the number of fund managers in the fund manager set to be preprocessed by the total number of the factor initial scores corresponding to the factors of the sample to be extracted in the sorted fund manager factor initial score set, which is equal to the sample extraction ratio.

For example, the sample extraction proportion is 40%, the first 40% of the factor initial scores are extracted from the sorted fund manager factor initial score set, and the fund managers corresponding to all the extracted factor initial scores are used as the fund manager set to be preprocessed, which is not specifically limited in this example.

And for S3125, respectively acquiring the fund data of the month to be extracted from the fund database for each fund manager in the fund manager set to be preprocessed, and taking all the acquired fund data as the fund data set to be analyzed.

Fund data is data for a fund over a month. Fund data includes, but is not limited to: fund identification, market value, monthly profitability.

And S3126, performing extremum removal on the fund manager set to be preprocessed according to the fund data set to be analyzed by adopting a median value removing method, and taking the fund manager set to be preprocessed of the fund manager from which the extremum is removed as the fund manager set of the sample to be extracted. Therefore, the influence of noise caused by the fund manager with extreme value on the factor verification accuracy is avoided, and the accuracy of the determined verification sample subset is improved.

The specific method steps for removing the extreme value from the fund manager set to be preprocessed by using a median value removing method according to the fund data set to be analyzed are not repeated herein.

And for S3127, fund data is acquired from the fund data set to be analyzed according to each fund manager in the fund manager set of the sample to be extracted, and all the acquired fund data is taken as the fund data set to be extracted. Thereby removing all fund data of the extremum fund manager.

And S3128, performing market value ratio calculation on each fund managed by the fund manager of the target sample to be extracted according to the fund data set to be extracted to obtain the market value ratio of the fund to be analyzed corresponding to each fund corresponding to the fund manager of the target sample to be extracted, wherein the fund manager of the target sample to be extracted is the fund manager of any one of the fund manager sets of the sample to be extracted.

For example, the fund manager of the target sample to be extracted manages 3 funds J1, J2, J3, the fund manager of the target sample to be extracted has a market value of S1, the fund manager of the target sample to be extracted has a market value of S2, and the fund manager of J3 has a market value of S3, the fund value occupation ratio to be analyzed corresponding to the fund J1 corresponding to the fund manager of the target sample to be extracted is (S1/(S1+ S2+ S3)), and the fund value occupation ratio to be analyzed corresponding to the fund J2 corresponding to the fund manager of the target sample to be extracted is (S2/(S1+ S2+ S3)), where the fund value occupation ratio to be analyzed corresponding to the fund J3 corresponding to the fund manager of the target sample to be extracted is (S3/(S1+ S2+ S3)), which is not specifically limited by this example.

For S3129, the to-be-analyzed fund market value ratio of the fund manager of the target to-be-extracted sample is used as a weight, the monthly profitability is weighted and summed for all funds managed by the fund manager of the target to-be-extracted sample according to the to-be-extracted fund data set, and the data obtained by the weighted summation is used as the monthly profitability comprehensive value of the fund manager corresponding to the fund manager of the target to-be-extracted sample, where the to-be-analyzed fund market value ratio corresponding to the fund J1 corresponding to the fund manager of the target to-be-extracted sample is (S3/(S1+ S2+ S3)), which is not specifically limited in this example.

And for S31210, generating verification samples according to the monthly profitability comprehensive values of the fund managers, and taking all the generated verification samples as the verification sample subset of the months to be extracted corresponding to the factors of the samples to be extracted.

For S31211, repeating steps S3122 through S31211 until the obtaining of the factors in the factor library is completed.

In an embodiment, the step of verifying each factor in the valid factor set according to the shanghai depth 300 index list, the verification sample set, the validity index set and the monotonicity index set to obtain a verification factor set includes:

s321: acquiring a factor from the effective factor set as a factor to be verified;

s322: obtaining a verification sample from the verification sample set according to the factor to be verified to obtain a verification sample set to be processed;

s323: acquiring a preset single validation rule of the validity index, and respectively performing validity validation on each validity index in the validity index set according to the preset single validation rule of the validity index, the Shanghai depth 300 index list and the to-be-processed validation sample set to obtain a validity index validation result set;

s324: acquiring a preset comprehensive validation rule of the validity index, and performing comprehensive validation on validity according to the preset comprehensive validation rule of the validity index and the validity index validation result set to obtain a comprehensive validation result;

s325: acquiring a preset monotonicity index single item verification rule, and respectively performing monotonicity verification on each monotonicity index in the monotonicity index set according to the preset monotonicity index single item verification rule, the Shanghai depth 300 index list and the to-be-processed verification sample set to obtain a monotonicity index verification result set;

s326: acquiring a preset monotonicity index comprehensive verification rule, and performing monotonicity comprehensive verification according to the preset monotonicity index comprehensive verification rule and the monotonicity index verification result set to obtain a monotonicity comprehensive verification result;

s327: when the validity comprehensive verification result and the monotonicity comprehensive verification result both pass, determining the factor to be verified as a verification passing factor;

s328: repeatedly executing the step of acquiring a factor from the effective factor set as a factor to be verified until all the factors in the effective factor set are acquired;

s329: and taking all the factors passing the verification as the verification factor set.

In this embodiment, each factor in the effective factor set is verified according to the Shanghai depth 300 index list, the verification sample set, the validity index set, and the monotonicity index set, so that the validity verification and the monotonicity verification are satisfied when the contract of the factor set is verified, and thus, a factor that accurately expresses a fund manager is found, which is beneficial to further improving the accuracy of the fund manager recommendation.

For S321, sequentially obtaining a factor from the valid factor set, and taking the obtained factor as a factor to be verified.

For step S322, all the verification samples corresponding to the factor to be verified are obtained from the verification sample set, and all the obtained verification samples are used as the verification sample set to be processed. That is, the verification samples in the set of verification samples to be processed are all the verification samples corresponding to the factor to be verified.

For S323, the validity verification of each validity index is performed on the validity index set according to the preset validity index single-item verification rule, the Shanghai depth 300 index list, and the to-be-processed verification sample set, where the number of validity index verification results in the validity index verification result set is the same as the number of validity indexes in the validity index set.

The preset validity index single item verification rule comprises a threshold range corresponding to each validity index, wherein the threshold range corresponding to the validity index means that the validity of the validity index passes verification.

And the effectiveness index is used for tracking the combined performance of the to-be-processed verification sample set to investigate the effectiveness of the factor.

The validity indicators in the set of validity indicators include, but are not limited to: one or more of monthly profit averages, information ratios, combined wins, standard deviations, cumulative profitability, and single-month maximum fall. And the monthly income average value is obtained by carrying out average calculation on the monthly income comprehensive values of all fund managers in the to-be-processed verification sample set. The information ratio, i.e., the information ratio, refers to the risk adjustment excess for the validation sample set to be processed over the same month's Shanghai depth 300 index (data from the list of Shanghai depth 300 indices). The combined win ratio refers to the number of the verification samples of the verification sample set to be processed, of which the fund manager monthly yield comprehensive value is higher than the Shanghai depth 300 index in the same month, divided by the number of the verification samples of the verification sample set to be processed. And the sum of the monthly profitability comprehensive values of all fund managers in the to-be-processed verification sample set. The maximum fall per month is the minimum value of each month in the monthly profitability comprehensive values of each fund manager in the verification sample set to be processed.

For S324, the preset validity index integrated verification rule is all passed.

Optionally, when all the validity index verification results in the validity index verification result set pass, it is determined that the validity comprehensive verification result passes, otherwise, it is determined that the validity comprehensive verification result fails.

For S325, performing monotonicity verification of each monotonicity index on the monotonicity index set according to the preset monotonicity index single verification rule, the Shanghai depth 300 index list, and the to-be-processed verification sample set, wherein the number of monotonicity index verification results in the monotonicity index verification result set is the same as the number of monotonicity indexes in the monotonicity index set.

The preset monotonicity index single item verification rule comprises a threshold range corresponding to each monotonicity index, wherein the threshold range corresponding to the monotonicity index means that the validity of the monotonicity index is verified to be passed.

The monotonicity index examines the effectiveness of the factor by analyzing whether the performance of the fund managed by the fund manager of each fund manager grading has monotonicity.

The monotonicity indicators in the monotonicity indicator set include, but are not limited to: the accumulated earning rate of each grade, the accumulated earning rate of each grade relative to a reference, the average annual profit rate of each grade and the average annual profit rate of each grade relative to a reference. The accumulated yield of each grade is the accumulated yield of each fund manager in grade. The accumulated yield of each grade relative to the benchmark is the accumulated yield of each fund manager grade relative to the benchmark, wherein the benchmark refers to the Shanghai depth 300 index of the same month (the data is derived from the Shanghai depth 300 index list). The average annual profit rate of each grade refers to the average annual profit rate of each fund manager. The average annual benefit rate of each grade relative to the benchmark refers to the average annual benefit rate of each fund manager grade relative to the benchmark, wherein the benchmark refers to the Shanghai depth 300 index of the same month (the data is from the Shanghai depth 300 index list).

For S326, the preset monotonicity index integrated verification rules are all passed.

Optionally, when all the monotonicity index verification results in the monotonicity index verification result set pass, determining that the monotonicity comprehensive verification result passes, otherwise, determining that the monotonicity comprehensive verification result does not pass.

For S327, when both the validity comprehensive verification result and the monotonicity comprehensive verification result pass, it means that the factor to be verified passes validity and monotonicity verification, and the factor to be verified is closely related to performance of the fund manager, so that the factor to be verified may be determined as a factor that passes verification.

For step S328, the step of obtaining a factor from the valid factor set as the factor to be verified is repeated until the obtaining of all the factors in the valid factor set is completed

For S329, all the factors that pass the verification are taken as a set, and the set is taken as the set of verification factors.

Referring to fig. 2, the present application also proposes a recommendation apparatus for a fund manager, the apparatus comprising:

the data acquisition module 100 is configured to acquire a fund manager factor standard score set of historical adjacent months of the target recommended month;

the effective factor collection obtaining module 200 is configured to perform effective factor screening according to the fund manager factor standard score collection by using a relevance comparison method between the grading ranking and the average profitability ranking to obtain an effective factor collection;

and a recommendation fund manager set determining module 300, configured to obtain a preset factor combination rule, and perform fund manager recommendation according to the effective factor set and the fund manager factor standard score set by using the preset factor combination rule to obtain a recommendation fund manager set.

In the embodiment, the fund manager recommendation is performed from multiple factor dimensions by obtaining the fund manager factor standard score set of historical adjacent months of the target recommended month, performing effective factor screening according to the fund manager factor standard score set by using a correlation comparison method of a grading ranking and an average profitability ranking, obtaining the effective factor set, obtaining the preset factor combination rule, and performing the fund manager recommendation according to the effective factor set and the fund manager factor standard score set by using the preset factor combination rule, so that the fund manager recommendation is performed in full consideration of the investment performance and other aspects of the fund manager, and the recommendation accuracy of the fund manager is improved.

Referring to fig. 3, a computer device, which may be a server and whose internal structure may be as shown in fig. 3, is also provided in the embodiment of the present application. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer designed processor is used to provide computational and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used for storing data such as recommendation methods of fund managers and the like. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a recommendation method for a fund manager. The recommendation method of the fund manager comprises the following steps: acquiring a fund manager factor standard score set of historical adjacent months of a target recommended month; screening effective factors according to the fund manager factor standard score set by adopting a relevance comparison method of grading ranking and average profitability ranking to obtain an effective factor set; and acquiring a preset factor combination rule, and recommending the fund managers according to the effective factor set and the fund manager factor standard scoring set by adopting the preset factor combination rule to obtain a recommended fund manager set.

In the embodiment, the fund manager recommendation is performed from multiple factor dimensions by obtaining the fund manager factor standard score set of historical adjacent months of the target recommended month, performing effective factor screening according to the fund manager factor standard score set by using a correlation comparison method of a grading ranking and an average profitability ranking, obtaining the effective factor set, obtaining the preset factor combination rule, and performing the fund manager recommendation according to the effective factor set and the fund manager factor standard score set by using the preset factor combination rule, so that the fund manager recommendation is performed in full consideration of the investment performance and other aspects of the fund manager, and the recommendation accuracy of the fund manager is improved.

An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, the computer program, when executed by a processor, implementing a recommendation method for a fund manager, including the steps of: acquiring a fund manager factor standard score set of historical adjacent months of a target recommended month; screening effective factors according to the fund manager factor standard score set by adopting a relevance comparison method of grading ranking and average profitability ranking to obtain an effective factor set; and acquiring a preset factor combination rule, and recommending the fund managers according to the effective factor set and the fund manager factor standard scoring set by adopting the preset factor combination rule to obtain a recommended fund manager set.

In the recommendation method for the executed fund manager, the effective factor set is obtained by obtaining the fund manager factor standard score set of the historical adjacent month of the target recommended month and performing effective factor screening according to the fund manager factor standard score set by using a correlation comparison method of the grading ranking and the average profitability ranking, the preset factor combination rule is obtained, the fund manager is recommended according to the effective factor set and the fund manager factor standard score set by using the preset factor combination rule to obtain the recommended fund manager set, and the fund manager recommendation is performed from multiple factor dimensions, so that the investment performance and other features of the fund manager are fully considered, and the recommendation accuracy of the fund manager is improved.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium provided herein and used in the examples may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double-rate SDRAM (SSRSDRAM), Enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and bus dynamic RAM (RDRAM).

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.

The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are also included in the scope of the present application.