Pension product screening methods and devices

Abstract

The present invention provides a pension product screening method and device, which can be used in the field of financial technology, the method comprising: obtaining investment data of pension products; analyzing the strategy style labels of pension products according to the investment data; analyzing the attribute labels of pension products with different strategy style labels; adding pension products that meet the white list access requirements to a candidate pool according to the white list access requirements corresponding to different strategy style labels; querying the pension product blacklist, and removing the pension products in the candidate pool that belong to the pension product blacklist; determining the proportion of pension products with different strategy style labels under different attribute labels according to the market view data of the investment manager, and adding the corresponding proportion of pension products in the candidate pool to the core pool; outputting the candidate pool and the core pool. The present invention can be used to classify pension products.

Description

Technical Field

[0001] The present invention relates to the field of financial technology, and in particular to a pension product screening method and device. Background Art

[0002] When selecting and evaluating pension products, identifying the investment characteristics of pension products can help investors accurately grasp the product characteristics and choose appropriate pension products under appropriate market conditions. However, how to identify the investment characteristics of pension products in order to classify pension products is an issue that needs to be addressed urgently. Summary of the invention

[0003] The embodiment of the present invention provides a pension product screening method for classifying pension products. The method includes:

[0004] Obtain investment data for pension products;

[0005] Analyze the strategy style labels of pension products based on the investment data;

[0006] Analyze the attribute labels of pension products with different strategy style labels;

[0007] According to the whitelist access requirements corresponding to different strategy style labels, pension products that meet the whitelist access requirements are added to the candidate pool;

[0008] Check the pension product blacklist and remove the pension products on the pension product blacklist from the candidate pool;

[0009] According to the market view data of the investment manager, determine the proportion of pension products with different strategy style labels under different attribute labels, and add the corresponding proportion of pension products in the alternative pool to the core pool;

[0010] Output the candidate pool and core pool.

[0011] The embodiment of the present invention provides a pension product screening device for classifying pension products. The device includes:

[0012] Investment data acquisition module, used to obtain investment data of pension products;

[0013] A strategy style label analysis module, used to analyze the strategy style label of the pension product based on the investment data;

[0014] Attribute label analysis module, used to analyze the attribute labels of pension products with different strategy style labels;

[0015] The candidate pool preliminary selection module is used to add pension products that meet the whitelist access requirements to the candidate pool according to the whitelist access requirements corresponding to different strategy style labels;

[0016] The candidate pool screening module is used to query the pension product blacklist and remove the pension products on the pension product blacklist from the candidate pool;

[0017] The core pool determination module is used to determine the proportion of pension products with different strategy style labels under different attribute labels based on the market view data of the investment manager, and add the corresponding proportion of pension products in the alternative pool to the core pool;

[0018] Output module, used to output the candidate pool and core pool.

[0019] An embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned pension product screening method when executing the computer program.

[0020] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned pension product screening method is implemented.

[0021] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned pension product screening method is implemented.

[0022] In an embodiment of the present invention, the investment data of pension products are obtained; based on the investment data, the strategy style labels of pension products are analyzed; the attribute labels of pension products with different strategy style labels are analyzed; based on the whitelist access requirements corresponding to different strategy style labels, pension products that meet the whitelist access requirements are added to the candidate pool; the pension product blacklist is queried, and the pension products in the candidate pool that are on the pension product blacklist are removed; based on the market view data of the investment manager, the proportion of pension products with different strategy style labels under different attribute labels is determined, and the pension products with corresponding proportions in the candidate pool are added to the core pool; the candidate pool and the core pool are output. In the above process, multiple types of pension products are progressively screened into the candidate pool and the core pool, thereby providing a more accurate performance classification, which is convenient for subsequent key tracking and analysis of the core pool products. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0024] Figure 1 The process of the pension product screening method in the embodiment of the present invention is as follows Figure 1 ;

[0025] Figure 2 It is a schematic diagram of the pension product screening in the implementation of the present invention;

[0026] Figure 3 is a calculation flow chart of the strategy style analysis model in an embodiment of the present invention;

[0027] Figure 4 The process of the pension product screening method in the embodiment of the present invention is as follows Figure 2 ;

[0028] Figure 5 is a calculation flow chart of the risk management model in an embodiment of the present invention;

[0029] Figure 6 A flowchart of determining the early warning level of a bond in an embodiment of the present invention;

[0030] Figure 7 This is a flow chart of determining whether the pension product is a high-risk warning pension product in an embodiment of the present invention;

[0031] Figure 8 This is a schematic diagram showing an interface in an embodiment of the present invention;

[0032] Fig. 9 Schematic diagram of a pension product screening device in an embodiment of the present invention Figure 1 ;

[0033] Fig.10 Schematic diagram of a pension product screening device in an embodiment of the present invention Figure 2 ;

[0034] Fig.11 Schematic diagram of a pension product screening device in an embodiment of the present invention Figure 3 ;

[0035] Fig.12 Schematic diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0037] In the description of this specification, the terms "include", "including", "have", "contain", etc. are all open terms, which mean including but not limited to. The descriptions with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. The order of steps involved in each embodiment is used to schematically illustrate the implementation of the present application, and the order of steps is not limited and can be appropriately adjusted as needed.

[0038] Figure 1 The process of the pension product screening method in the embodiment of the present invention is as follows Figure 1 ,like Figure 1 As shown, including:

[0039] Step 101, obtaining investment data of pension products;

[0040] Step 102, analyzing the strategy style label of the pension product according to the investment data;

[0041] Step 103, analyzing the attribute labels of pension products with different strategy style labels;

[0042] Step 104, according to the whitelist access requirements corresponding to different strategy style labels, add pension products that meet the whitelist access requirements to the candidate pool;

[0043] Step 105, query the pension product blacklist, and remove the pension products in the candidate pool that are on the pension product blacklist;

[0044] Step 106, according to the market view data of the investment manager, determine the proportion of pension products with different strategy style labels under different attribute labels, and add the corresponding proportion of pension products in the alternative pool to the core pool;

[0045] Step 107, outputting the candidate pool and the core pool.

[0046] Figure 2 This is a schematic diagram of the pension product screening in the implementation of the present invention. Figure 2 and Figure 1Correspondingly, each step is analyzed in detail below.

[0047] In step 101, investment data of pension products is obtained.

[0048] In one embodiment, the investment data includes at least one of the following data:

[0049] Basic information, net value and size data, asset distribution data, heavy holdings data, and market public data.

[0050] In one embodiment, obtaining investment data of a pension product includes:

[0051] The investment data of pension products from different sources are merged, deduplicated and completed.

[0052] In the above embodiment, the investment data sources of the pension product may include investor sources and market data supplier sources, wherein the investment data from the investor source includes asset distribution data, heavy position data, basic information, net value and scale data, and the investment data from the market data supplier source includes basic information, net value and scale data, and market public data. It can be seen that the investment data from the two sources contain duplicate data. Therefore, before using the investment data, it is necessary to first perform merging processing, deduplication processing and completion processing, wherein the supplementary processing is mainly to complete the data with incomplete identification such as date in the data. All investment data can be put into the pension product database for subsequent reading. In addition, basic information, net value and scale data, asset distribution data, and heavy position data can also be called basic data. Pay attention to the following two types of data when processing.

[0053] Basic information: Compare and summarize the basic information of pension products from different sources, compare them according to the information update time, and keep the basic information update records of each pension product. In the event of a conflict between the basic information of the investment manager and the market data supplier, the basic information of the investment manager shall prevail.

[0054] Net value and scale data: The net value and scale data of pension products from different sources are aggregated, and the daily rate of return of pension products is calculated based on the daily unit net value, and the missing scale data is filled in with the difference. In case of conflict, the data of the investment manager shall prevail. The data between the two published scale data dates are filled in with the previous scale data. For example, if the scale data published on January 5, 2020 is A, and the scale is changed to B on January 10, 2020, then the scale of the trading days from January 6 to January 9, 2020 is A.

[0055] In step 102, the strategy style label of the pension product is analyzed based on the investment data.

[0056] In one embodiment, the strategy style label includes three levels, wherein:

[0057] The primary strategy styles include absolute return and relative return;

[0058] The secondary strategy styles under the absolute return type include stock type, mixed type, currency type, market value pricing type, and cost pricing type; the tertiary strategy styles under the market value pricing type include convertible bond type, allocation tool type, and ordinary type; the tertiary strategy styles under the cost pricing type include debt pool type, single debt plan type, trust pool type, single trust type, and deposit type;

[0059] The secondary strategy styles under the relative return type include stock type and mixed type; the tertiary strategy styles under the stock type include Hong Kong stocks type, specific industry type, index enhancement type, flexible allocation type, value large-cap type, value small-cap type, growth large-cap type, and growth small-cap type; the tertiary strategy styles under the mixed type include allocation tool type, aggressive type, and fixed income + tool type.

[0060] In one embodiment, analyzing the strategy style label of the pension product according to the investment data includes:

[0061] The investment data is input into a strategy style analysis model to obtain a strategy style label for the pension product.

[0062] The style strategy analysis model can calculate investment data and then derive the three levels of strategy style labels for each pension product. Table 1 shows the classification of strategy style labels for pension products.

[0063] Table 1

[0064]

[0065]

[0066] When indicating the strategy style label, “-” can be used. For example, a certain pension product belongs to the absolute return type-stock type. For another example, a certain pension product belongs to the absolute return type-market value valuation type-allocation tool type.

[0067] The following is a detailed introduction to the strategy style labels for each level.

[0068] C1. Absolute return type

[0069] C1.1 Absolute Return Type - Stock Type

[0070] For stock-based pension products registered with the regulatory authorities, within the measurement time interval, if the monthly asset distribution positions exclude the equity positions hedged by stock index futures, that is, the average equity long exposure is less than the threshold μ1, then it is an absolute return-stock-based pension product. Among them, the monthly asset distribution positions and the equity positions hedged by stock index futures can be obtained from the asset distribution data.

[0071] C1.2 Absolute return type - mixed type

[0072] If the maximum drawdown of a hybrid pension product registered with the regulatory authority in the past three years is less than the threshold μ2, it is an absolute return hybrid pension product. The maximum drawdown in the past three years can be calculated based on the net asset value time series data, and the net asset value time series data can be obtained from public market data.

[0073] C1.3 Absolute Return Type - Currency Type

[0074] The monetary pension products registered with the regulatory authorities are all absolute return-monetary pension products.

[0075] C1.4.1 Absolute return type - market value type - convertible bond type

[0076] If the fixed-income pension products registered with the regulatory authorities have monthly asset distribution holdings and convertible bond positions exceeding the threshold μ3 during the measurement period, then they are absolute return-market-valued-convertible bond pension products. Among them, the monthly asset distribution holdings and convertible bond positions can be obtained from the asset distribution data.

[0077] C1.4.2 Absolute return type - market value type - configuration tool type

[0078] If the maximum annual drawdown of a fixed-income pension product registered with the regulatory authority since its establishment is less than the threshold μ4, then it is an absolute return-market value-based-allocation tool-type pension product, or the investment manager determines that the pension product is an absolute return-market value-based-allocation tool-type product through reporting.

[0079] C1.4.3 Absolute return type - market value type - ordinary type

[0080] General fixed-income pension products registered with regulatory authorities, and non-convertible bonds and allocation tool-type pension products are all absolute return-market value-based general pension products.

[0081] C1.5.1 Absolute return type - cost pricing type - debt pool type

[0082] If a creditor's pension product is registered with the regulatory authority and there is no single non-standard project with a position greater than the threshold μ5, it is an absolute return-cost-based-creditor's pool pension product. Among them, the holding of non-standard projects belongs to the registration information of pension products and is submitted by the investment manager.

[0083] C1.5.2 Absolute return type - cost pricing type - single debt plan type

[0084] If a creditor's pension product is registered with the regulatory authority and the investment position of a single non-standard project is greater than the threshold μ6, it is an absolute return-cost-based-single creditor's plan pension product. Among them, the investment position of a single non-standard project belongs to the registration information of the pension product, which shall be submitted by the investment manager.

[0085] C1.5.3 Absolute return type - cost pricing type - trust pool type

[0086] The regulatory agency has registered trust pension products and no holdings are greater than the threshold μ 71 The individual non-standard projects are absolute return-cost-based trust pool pension products.

[0087] C1.5.4 Absolute return type - cost pricing type - single trust type

[0088] If a trust-type pension product is registered with the regulatory authority and the investment position in a single non-standard project is greater than the threshold μ8, it is an absolute return-cost-based single trust-type pension product.

[0089] C1.5.5 Absolute return type - cost pricing type - deposit type

[0090] The deposit-type pension products registered with the regulatory authorities are all absolute return-cost-based deposit-type pension products.

[0091] C2. Relative return type

[0092] C2.1.1 Relative return type - stock type - Hong Kong stock type

[0093] The Hong Kong stock-type pension products registered with the regulatory authorities are all relative return-equity-Hong Kong stock-type pension products.

[0094] C2.1.2 Relative return type - stock type - specific industry type

[0095] Relative return type - stock type - specific industry type refers to the stock type pension products registered with the regulatory agency whose actual investment field is limited to a certain industry sector. It is calculated by measuring the beta value of the pension product and the Shenwan Industry Level 1 Industry Index during the period:

[0096]

[0097] Among them, R is the daily rate of return of the pension product, is the daily return rate of Shenwan first-level industry index i, ε i is the residual; through the regression within the measurement period, we can get the alpha value α of the pension product relative to different Shenwan first-level industries i and the systematic risk factor β i If MAX(β i )>μ7, the investment style characteristics of the pension product are relative return type-stock type-specific industry type, and the specific industry is the Shenwan first-level industry corresponding to the maximum β value.

[0098] C2.1.3 Relative return type - stock type - index enhancement type

[0099] Relative return-stock-index-enhanced refers to stock-based pension products registered with the regulatory agency that actually invest and track a specific index. The beta' value of the pension product and the market index over time is calculated:

[0100]

[0101] Among them, R' is the daily rate of return of pension products, is the daily return of market index j, ε j ′ is the residual; by regressing within the measurement period, we can obtain the alpha value α of the pension product relative to different market indices j ′ and systematic risk coefficient β j ′. If MAX(β j ′)>μ8, the investment style of the pension product is relative return-stock-index enhanced, and the specific index is the market index corresponding to the maximum β′ value.

[0102] C2.1.4-C2.1.8 Relative return type - stock type

[0103] Within the measurement time interval, according to the set time window length T, firstly, the stock-based pension product styles registered with the regulatory agency within each time window are measured as follows:

[0104]

[0105] where r i is the daily rate of return of pension products, y j,i is the daily return sequence of the national equity style index j, c j is the correlation coefficient between the pension product and the corresponding national securities style index j. The national securities style indexes are the national securities large-cap growth index, the national securities small-cap growth index, the national securities large-cap value index and the national securities small-cap value index. The style of the pension product in this time window is c jThe styles corresponding to the maximum values ​​are Large-Cap Growth, Small-Cap Growth, Large-Cap Value, and Small-Cap Value.

[0106] Then, the stability of the pension product style in each time window within the measurement time interval is judged. The measurement time interval is divided into N time windows, and the occurrence frequencies of the four styles are C j ,like

[0107]

[0108] That is, the frequency of the style with the highest frequency is not less than the set threshold The style of the pension product is the most frequent strategy style label, which is Growth Large Cap, Growth Small Cap, Value Large Cap or Value Small Cap. Otherwise, the style of the pension product in each time window is not stable, and the strategy style label of the pension product is Flexible Allocation.

[0109] C2.2 Relative return type - mixed type

[0110] During the measurement period, the monthly asset distribution of the hybrid pension products registered with the regulatory agency is determined, excluding the equity positions hedged by stock index futures, that is, the average equity long exposure. like It is a relative income-mixed-aggressive pension product; if It is a relative income-mixed-allocation tool pension product; if It is a relative return-mixed-fixed income + tool-type pension product.

[0111] Figure 3 : is a calculation flow chart of the strategy style analysis model in an embodiment of the present invention. In one embodiment, the calculation steps of the strategy style analysis model include:

[0112] Step 301, querying the registration status of the regulatory agency of the pension product and determining the strategy style label of the pension product;

[0113] Step 302, when the strategy style label of the pension product cannot be determined according to the record status of the regulatory agency, the strategy style label of the pension product is determined according to the calculation method of the strategy style labels of various pension products.

[0114] In Table 1, corresponding supplementary tags can be manually added to facilitate understanding of the strategy style tags when displayed later. In step 103, attribute tags of pension products with different strategy style tags are analyzed.

[0115] In one embodiment, the attribute labels include one or any combination of excess return type, market type, manager performance type, net asset size type, heavily invested industry type, heavily invested stock type, time-weighted annualized rate of return ranking type, time-weighted annualized rate of return ranking stability type, drawdown type, duration type, leverage type, credit type, and turnover rate type.

[0116] Among them, after analyzing the strategy style labels of pension products, multiple pension products are grouped according to attribute labels under each strategy style label.

[0117] For example, determine the type of excess return for each pension product, and then group all pension products according to the type of excess return under each strategy style label.

[0118] Table 2 is an example of adding attribute label grouping and supplementary labels to Table 1. The above attribute labels are classified into general attribute labels and specific type attribute labels.

[0119] Table 2

[0120]

[0121]

[0122] The following is a detailed introduction to each attribute tag.

[0123] S1. Excess return type

[0124] The types of excess returns specifically include high-return IR and low-return IR. Among them, pension products whose excess return IR ranking (in positive order) is in the top p1% of the overall quantile are high-return IR type; pension products whose ranking is in the bottom p2% of the overall quantile are low-return IR type. The calculation formula of excess return IR is as follows:

[0125]

[0126] Among them, the numerator is the average value of the product's excess return, r c is the excess return of the product, r j is the excess return of the benchmark, and the denominator is the standard deviation of the excess return of the product. The benchmarks of different types of products are different.

[0127] The benchmark for stock-based funds is: 100% Shanghai and Shenzhen 300 Index;

[0128] The benchmark for the hybrid type is: a% Shanghai and Shenzhen 300 Index + (100% - a%) China Bond New Comprehensive Wealth Index;

[0129] The benchmarks for fixed income and money market types are: 100% China Bond New Comprehensive Wealth Index.

[0130] S2. Market Type

[0131] Market types specifically include bull market performance, bear market performance, volatile market performance, bull market poor performance, bear market poor performance, and volatile market poor performance. Market types are market types based on the stock market, bond market, and convertible bond market. The annualized yields of pension products under different market types are calculated and ranked. The top p3% of the overall quantiles are the performance-based products of the corresponding market types, and the bottom p4% of the overall quantiles are the poor performance products of the corresponding market types. The calculation of the annualized yield requires the use of a market representative index. Different pension products are suitable for different measurement markets and market representative indexes:

[0132] Stock type and mixed type: the market to be used is the stock market, and the market representative index is the Shanghai and Shenzhen 300 Index;

[0133] Fixed income type (non-convertible bond type), currency type: the market to be used is the bond market, and the market representative index is the China Bond New Comprehensive Wealth Index;

[0134] Convertible bond type: The market to be used is the convertible bond market, and the market representative index is the CSI Convertible Bond Index.

[0135] S3. Manager Performance Type

[0136] The pension products with the same strategy style label managed by the manager are aggregated to form the manager's net value and yield curve, and the time-weighted rate of return of the manager in the measured time period is calculated. Managers whose time-weighted rate of return in the measured time period ranks in the top p5% of the overall percentile are considered to be high-performing managers, and those whose time-weighted rate of return ranks in the bottom p6% of the overall percentile are considered to be poor-performing managers.

[0137] S4. Net worth scale type

[0138] Pension products whose net asset value scale is in the top p7% of the overall percentile are large-scale products, while those in the bottom p8% of the overall percentile are small-scale products.

[0139] S5. Types of heavily invested industries

[0140] The top N Shenwan first-level industries held by equity pension products will have corresponding N industry labels added to the pension products, marked as heavily-weighted industries*.

[0141] S6. Types of heavily-weighted stocks

[0142] The top N stocks held by stock-based pension products, if there are stocks whose net value position is greater than the threshold Add corresponding individual stock name labels to pension products and mark them as heavily-weighted stocks*.

[0143] S7. Time-weighted annualized rate of return ranking type

[0144] In the measurement time interval, according to the set time window length T, firstly, the time-weighted annualized rate of return of pension products in each time window is calculated and ranked, and then the ranking of all time windows is averaged. Pension products with the average ranking in the top p9% are high-ranking products, and those in the bottom p 10 % of pension products are low-ranking.

[0145] S8. Time-weighted annualized rate of return ranking stability type

[0146] In the measurement time interval, according to the set time window length T, first calculate and rank the time-weighted annualized rate of return of pension products in each time window, and then calculate the standard deviation of the ranking of all time windows. The ranking standard deviation (in positive order) is in the top p 11 % of pension products are of stable ranking, ranked at the bottom 12 % of pension products are of the ranking fluctuation type.

[0147] S9. Retracement Type

[0148] The maximum drawdown ranking (absolute value positive order) is in the top p of the overall quantile 13 %, it is a low retracement type; the ranking is after the overall percentile p 14 %, it is a high retracement type.

[0149] S10. Duration Type

[0150] The weighted duration ranking (positive order) of bonds held by fixed-income pension products is in the top p of the overall quantile 15 %, it is a short-duration type; the ranking is after the overall quantile p 16 %, it is a long-term type.

[0151] S11. Leverage Type

[0152] The positive repurchase position of the pension product holding bonds, i.e. the leverage ranking (positive order), is in the top p of the overall quantile 17 %, it is a high leverage type; the ranking is at the bottom of the overall percentile p 18 %, it is a low leverage type.

[0153] S12. Credit Type

[0154] The proportion of credit bond positions held by fixed-income pension products is greater than the threshold The credit label judgment is then performed. The position of credit bonds with a rating of no less than AAA is greater than p 19 %, it is a high credit type; the position of credit bonds with a bond rating not higher than AA is greater than p 20%, it is a low credit type.

[0155] S13. Turnover rate type

[0156] Measures the turnover rate of heavily invested products in pension products. Based on the monthly asset distribution data, the denominator is the frequency of heavily invested products filled in during the measurement period, and the numerator is the number of different heavily invested products filled in during the measurement period. The turnover rate is higher than the threshold p 21 % is a high turnover type, which is lower than the threshold p 22 % is a low turnover rate type.

[0157] In step 104, according to the whitelist access requirements corresponding to different strategy style tags, pension products that meet the whitelist access requirements are added to the candidate pool.

[0158] That is, different strategy style labels have different whitelist entry requirements, so as to select pension products with excellent overall performance to enter the whitelist, and the whitelist can be updated regularly. Products are ranked according to the same strategy style, and the index measurement interval is from the start date of the current investment manager of the pension product to the current evaluation date. The whitelist entry requirements are shown in Table 3.

[0159] Table 3

[0160]

[0161]

[0162] The scoring rules are mentioned in Table 3. In the scoring rules, the targeted pension products need to have been established for at least 6 months. Table 4 gives the corresponding scoring rules for stock-based pension products.

[0163] Table 4

[0164]

[0165] Table 5 gives the scoring rules for market value-based (including general and mixed) pension products.

[0166] Table 5

[0167]

[0168] In step 105, the pension product blacklist is queried, and the pension products in the candidate pool that are on the pension product blacklist are removed.

[0169] Figure 4 The process of the pension product screening method in the embodiment of the present invention is as follows Figure 2 In one embodiment, the method further comprises:

[0170] Step 401, input the pension product into the risk management model, determine whether the pension product is a high-risk warning pension product, and put the high-risk warning pension product into the blacklist.

[0171] Figure 5 The following is a calculation flow chart of the risk management model in an embodiment of the present invention. In one embodiment, the calculation steps of the risk management model include:

[0172] Step 501, based on the warning conditions of the bond, when it is determined that the bond is a warning bond, obtaining the warning level of the warning bond;

[0173] Step 502: When the pension product is a standard type, determine whether the pension product is a high-risk warning pension product according to the warning level of the warning bond;

[0174] Step 503, when the pension product is non-standard, when the net value of the pension product is adjusted downward or the daily rate of return is negative, determine that the pension product is a high-risk warning pension product;

[0175] Step 504: after determining that the pension product is determined to be a high-risk warning pension product, query the associated warning pension products of the pension product to determine that the associated warning pension product is a high-risk warning pension product.

[0176] Among them, the standard type refers to pension products whose primary strategy style is absolute return and whose secondary strategy style is market value-based; and the primary strategy style is relative return and whose secondary strategy style is a mixed pension product.

[0177] Figure 6 The flowchart of determining the warning level of a bond in an embodiment of the present invention is as follows. In one embodiment, based on the warning conditions of the bond, when it is determined that the bond is a warning bond, obtaining the warning level of the bond includes:

[0178] Step 601, calculating the value of each type of indicator for early warning of the bond;

[0179] Step 602, when the value of the indicator meets the warning condition, determine the indicator as a warning indicator, determine the warning level of the warning indicator, and determine the bond as a warning bond;

[0180] Step 603, determining the warning level of the warning bond according to the warning levels of all warning indicators of the warning bond.

[0181] In the above embodiment, the indicators used for early warning of bonds include the daily yield of the bond's trading price, the maximum retracement of the trading price in the first period, the deviation of the trading price from the China Bond Valuation, the daily yield of the China Bond Valuation, and the maximum retracement of the China Bond Valuation in the first period. Among them, the daily yield of the trading price and the maximum retracement of the trading price in the first period are used to calculate the volatility of the bond market trading price; the deviation of the trading price from the China Bond Valuation is used to calculate the deviation of the bond market trading price from the China Bond Valuation; the daily yield of the China Bond Valuation and the maximum retracement of the China Bond Valuation in the first period are used to calculate the volatility of the bond China Bond Valuation. The following is a detailed introduction.

[0182] Transaction price daily yield r: When the daily yield r < threshold t1, the bond triggers an early warning and the indicator is determined as an early warning indicator;

[0183] Maximum retracement m of the transaction price in the first period: When the absolute value of the maximum retracement m > threshold t2, the bond triggers an early warning and the indicator is determined as an early warning indicator;

[0184] The transaction price deviates from the China Bond Valuation by b: When the absolute value of the deviation from b is greater than the threshold t3, the bond triggers an early warning and the indicator is determined as an early warning indicator;

[0185] ChinaBond Valuation Daily Yield R: When the daily yield R < threshold t4, the bond triggers an early warning, and the indicator is determined as an early warning indicator;

[0186] Maximum drawdown M in the first period of Chinabond valuation: When the absolute value of maximum drawdown M > threshold t5, the bond triggers an early warning and the indicator is determined as a early warning indicator.

[0187] The bond's warning level is determined based on the warning levels of all warning indicators of the bond. For example, the daily yield of the trading price and the deviation of the trading price from the China Bond Valuation are warning indicators. The warning levels of the two warning indicators are 1 and 2 respectively. If other indicators do not trigger warnings, the bond's warning level is 3.

[0188] Figure 7 The flowchart of determining whether the pension product is a high-risk warning pension product in an embodiment of the present invention, in one embodiment, determining whether the pension product is a high-risk warning pension product according to the warning level of the warning bond includes:

[0189] Step 701, calculating the excess rate of return of the standard pension product;

[0190] Step 702, determining the yield of the warning bond according to the warning level of the warning bond;

[0191] Step 703, determining the warning triggering condition based on the yield of the warning bond and the benchmark yield;

[0192] Step 704: When the excess rate of return of the standard pension product meets the warning triggering condition, determine whether the pension product is a high-risk warning pension product.

[0193] Among them, the excess returns of standard pension products include the excess returns based on the pension product performance benchmark and the average excess returns of similar pension products.

[0194] In step 701, calculating the excess return based on the pension product performance benchmark includes:

[0195] Determine the performance benchmark of pension products based on the core asset classes held by each type of pension product;

[0196] Based on the performance benchmark of the pension product, calculate the excess return based on the performance benchmark of the pension product.

[0197] Among them, the major asset categories held by pension products are as follows:

[0198] Equity assets: CSI 300 Index;

[0199] Fixed income assets (excluding convertible bonds): ChinaBond New Comprehensive Wealth Index;

[0200] Convertible bonds: CSI Convertible Bond Index;

[0201] Monetary assets: one-year time deposit yield.

[0202] The performance benchmarks of pension products are determined based on the major asset classes held by various types of pension products, including:

[0203] For absolute return-market value-based pension products: Pension product performance benchmark = c% China Bond New Comprehensive Wealth Index + d% one-year fixed deposit;

[0204] For relative return-type mixed pension products: the performance benchmark of the pension product = e% Shanghai and Shenzhen 300 Index + f% China Bond New Comprehensive Wealth Index + g% one-year fixed deposit.

[0205] The above c, d, and g are numbers between 0 and 100.

[0206] Based on the performance benchmark of the pension product, calculate the excess return based on the performance benchmark of the pension product, including:

[0207] Calculate the rate of return of the performance benchmark of pension products;

[0208] The excess return based on the pension product performance benchmark = the rate of return of the pension product - the rate of return of the performance benchmark.

[0209] Among them, the yield can be calculated as daily yield or cumulative yield over the past N days (trading days).

[0210] Calculate the average excess return of similar pension products, including:

[0211] The average return of pension products of the same type is used as a reference benchmark, that is, the average excess return of pension products of the same type = the rate of return of pension products - the average rate of return of the same type. The rate of return can also be the daily rate of return or the cumulative rate of return over the past N days (trading days).

[0212] In step 703, the early warning triggering conditions need to be determined. For standard pension products such as relative return-hybrid and absolute return-market value-based pension products, the early warning triggering conditions are derived as follows:

[0213] The sensitivity requirement for absolute return-market value pension products is μ f , the sensitivity requirement for relative return-mixed pension products is μ h , that is, when the proportion of warning bonds held by absolute return-market value pension products exceeds μ f The risk management model needs to be able to identify.

[0214] Assuming that there are critical value products, the holdings of warning bonds with the required proportion of sensitivity are used as reference benchmarks. If the proportion of warning bonds held by pension products is too high or other assets lose too much, the product returns will theoretically exceed the critical value product returns.

[0215] Based on this, the pension product early warning trigger mechanism is established as follows:

[0216] r i -r s <r t -r s

[0217] Among them, r i is the pension product yield, r s is the benchmark rate of return, r t is the critical value product yield, that is:

[0218]

[0219] Among them, r alpha is the excess return of pension products, μ is the sensitivity requirement, The yield of the warning bond (the average yield of all warning bonds on the day or the average cumulative yield of all warning bonds in the past N days) can form the warning trigger condition:

[0220]

[0221] The two types of excess returns of pension products (daily return or cumulative return over the past N days) are monitored daily. When the daily return or the cumulative return over the past N days triggers the warning trigger conditions, a warning will be issued to the pension product to determine whether the pension product is a high-risk warning pension product.

[0222] In step 503, for fixed-income pension products other than the standard type, namely, absolute return-cost-based pension products and absolute return-monetary pension products, since the net value of non-standard pension products will be adjusted down when a credit crisis occurs in the assets held by them, when the net value of the pension product is adjusted down or the daily yield is negative, the pension product is determined to be a high-risk warning pension product.

[0223] In step 504, it is necessary to query the related warning pension products of the high-risk warning pension products. The embodiment of the present invention proposes to use Copula function calculation.

[0224] The estimated parameters of Gumbel Copula and Clayton Copula functions of two pension products of the same three-level type can be calculated respectively. The unit net value, daily rate of return or daily excess rate of return of pension products can be used as model variables. The sample is the data of N trading days in the sample time window pushed back from the calculation time:

[0225] Specifically, Gumbel Copula is used to characterize the upper tail correlation of pension products, where α is the parameter to be estimated, representing the degree of correlation between the two variables. When α approaches 0, the two variables u and v tend to be completely correlated:

[0226]

[0227] Clayton Copula is used to characterize the lower tail correlation of pension products, where θ is the parameter to be estimated, representing the degree of correlation between two variables. When θ approaches 1, the two variables u and v tend to be completely correlated:

[0228]

[0229] The tail correlation of two pension products of the same type is calculated based on the estimated parameters of the Copula function. The upper and lower tail correlations of pension products are:

[0230] λ U =2-2 -α

[0231]

[0232] According to the warning threshold μ D and μ U , the lower tail correlation coefficient λ DGreater than or equal to the threshold μ D And the upper tail correlation coefficient λ U Less than or equal to the threshold μ U Both pension products triggered warnings and are related warning pension products.

[0233] In step 106, the proportions of pension products with different strategy style labels under different attribute labels are determined based on the market view data of the investment manager, and the corresponding proportions of pension products in the alternative pool are added to the core pool.

[0234] Table 6 is an example of investment manager market sentiment data. The investment manager market sentiment data generated by the investment manager may be received periodically (eg, monthly).

[0235] Table 6

[0236]

[0237] As can be seen from Table 6, the investment manager market view data includes two major categories.

[0238] The first is the overall view, including optimism, neutrality and pessimism. The second is the operating style expectation. Therefore, there are two ways to determine the proportion of pension products with different strategy style labels under different attribute labels.

[0239] In one embodiment, the proportions of pension products with different strategy style labels under different attribute labels are determined based on the market view data of the investment manager, including:

[0240] The proportion of investment managers with optimistic, neutral and pessimistic views in the market opinion data of investment managers corresponding to pension products with each strategy style label is determined as the proportion of pension products with this strategy style label under different attribute labels.

[0241] For example, for a relative return-stock pension product, the number of investment managers with optimistic, neutral, and pessimistic views are a, b, and c respectively. Then, among the pension products with the attribute label of market type, the ratio of the number of bull market performance products, volatile market performance products, and bear market performance products is a:b:c.

[0242] In one embodiment, the proportions of pension products with different strategy style labels under different attribute labels are determined based on the market view data of the investment manager, including:

[0243] The proportion of investment managers with different operating styles expected in the investment manager market opinion data corresponding to the pension products with each strategy style label is determined as the proportion of the pension products with this strategy style label.

[0244] For example, for a relative return-stock pension product, the number of investment managers whose expected operating style is large-cap and small-cap are a and c respectively. Then, among the relative return-stock pension products, the ratio of the number of pension products whose three-level strategy styles involve large-cap and small-cap is a:c.

[0245] After determining the ratio, obtain the total number of core pool upper limit settings to determine the quantity of each pension product.

[0246] In one embodiment, a corresponding proportion of pension products in the alternative pool is added to the core pool, including:

[0247] Calculate the risk-return index value of each pension product in the candidate pool;

[0248] The pension products selected into the core pool are determined based on the proportion of pension products with each strategy style label under different attribute labels, as well as the risk-return indicator value of pension products with each strategy style label.

[0249] In the above, the number of pension products under each attribute label under each strategy style label can be obtained. Here, it is necessary to determine which pension products to select to enter the core pool. A ranking can be made based on the risk-return index value of the pension products, and the top-ranked pension products will enter the core pool.

[0250] Among them, the risk-return indicator of absolute return pension products can be expressed by annualized rate of return, and the risk-return indicator of relative return pension products can be expressed by the Sharpe ratio.

[0251] In one embodiment, the method further comprises:

[0252] Displays the strategy style labels and attribute labels, alternative pools and core pools of pension products.

[0253] In the embodiment of the present invention, it supports downloading the above data in EXCEL format and downloading the graphics drawn according to the data in JPG format. Figure 8 This is a schematic diagram of the interface display in the embodiment of the present invention. Among them, the query parameters can be selected from the deadline, product pool, parameter scheme, strategy style label (including primary strategy style, secondary strategy style, tertiary strategy style), supplementary label and attribute label; the deadline is selected to query the product pool applicable to the date; the product pool can be selected from the full product, alternative pool and core pool; the parameter scheme can be selected from the pre-configured scheme, including thresholds, performance benchmarks, etc.; both the strategy style label and the attribute label can be selected multiple times, and if no selection is made, all pension products will be queried.

[0254] The query results show the strategy style labels, supplementary labels and attribute labels calculated for each pension product, as well as the smallest product pool to which the product belongs (core pool < alternative pool < all products). Below the query results, the unit net value time series curve of the selected pension product in the query time interval is displayed. Below the unit net value time series curve is the cumulative excess yield curve of the selected pension product compared to the benchmark.

[0255] In summary, in the method proposed in the embodiment of the present invention, the investment data of pension products are obtained; based on the investment data, the strategy style labels of pension products are analyzed; the attribute labels of pension products with different strategy style labels are analyzed; according to the whitelist access requirements corresponding to different strategy style labels, the pension products that meet the whitelist access requirements are added to the alternative pool; the pension product blacklist is queried, and the pension products in the alternative pool that are on the pension product blacklist are removed; according to the market view data of the investment manager, the proportion of pension products with different strategy style labels under different attribute labels is determined, and the pension products with corresponding proportions in the alternative pool are added to the core pool; and the alternative pool and the core pool are output. In the above process, multiple types of pension products are progressively screened into the alternative pool and the core pool, thereby providing a more accurate performance classification, which is convenient for the subsequent key tracking and analysis of the core pool products.

[0256] The present invention also provides a pension product screening device, the principle of which is the same as that of the pension product screening method, and will not be described in detail here.

[0257] Fig. 9 Schematic diagram of a pension product screening device in an embodiment of the present invention Figure 1 ,like Fig. 9 As shown, including:

[0258] An investment data acquisition module 901 is used to acquire investment data of pension products;

[0259] A strategy style label analysis module 902 is used to analyze the strategy style label of the pension product according to the investment data;

[0260] The attribute label analysis module 903 is used to analyze the attribute labels of pension products with different strategy style labels;

[0261] The candidate pool preliminary selection module 904 is used to add pension products that meet the whitelist access requirements to the candidate pool according to the whitelist access requirements corresponding to different strategy style labels;

[0262] The candidate pool screening module 905 is used to query the pension product blacklist and remove the pension products in the candidate pool that are on the pension product blacklist;

[0263] The core pool determination module 906 is used to determine the proportion of pension products with different strategy style labels under different attribute labels according to the market view data of the investment manager, and add the pension products with corresponding proportions in the alternative pool to the core pool;

[0264] The output module 907 is used to output the candidate pool and the core pool.

[0265] In one embodiment, the investment data includes at least one of the following data:

[0266] Basic information, net value and size data, asset distribution data, heavy holdings data, and market public data.

[0267] In one embodiment, the investment data acquisition module is specifically used to:

[0268] The investment data of pension products from different sources are merged, deduplicated and completed.

[0269] In one embodiment, the strategy style label includes three levels, wherein:

[0270] The primary strategy styles include absolute return and relative return;

[0271] The secondary strategy styles under the absolute return type include stock type, mixed type, currency type, market value pricing type, and cost pricing type; the tertiary strategy styles under the market value pricing type include convertible bond type, allocation tool type, and ordinary type; the tertiary strategy styles under the cost pricing type include debt pool type, single debt plan type, trust pool type, single trust type, and deposit type;

[0272] The secondary strategy styles under the relative return type include stock type and mixed type; the tertiary strategy styles under the stock type include Hong Kong stocks type, specific industry type, index enhancement type, flexible allocation type, value large-cap type, value small-cap type, growth large-cap type, and growth small-cap type; the tertiary strategy styles under the mixed type include allocation tool type, aggressive type, and fixed income + tool type.

[0273] In one embodiment, the strategy style label analysis module is specifically used to:

[0274] The investment data is input into a strategy style analysis model to obtain a strategy style label for the pension product.

[0275] In one embodiment, the calculation steps of the strategy style analysis model include:

[0276] Check the record status of the pension product and determine the strategy style label of the pension product;

[0277] When the strategy style label of the pension product cannot be determined based on the filing situation, the strategy style label of the pension product is determined according to the calculation method of the strategy style labels of various types of pension products.

[0278] In one embodiment, the attribute labels include one or any combination of excess return type, market type, manager performance type, net asset size type, heavily invested industry type, heavily invested stock type, time-weighted annualized rate of return ranking type, time-weighted annualized rate of return ranking stability type, drawdown type, duration type, leverage type, credit type, and turnover rate type.

[0279] Fig.10 Schematic diagram of a pension product screening device in an embodiment of the present invention Figure 2 In one embodiment, the device further comprises a risk management module 1001, which is used to:

[0280] Input the pension product into the risk management model to determine whether the pension product is a high-risk warning pension product, and put the high-risk warning pension product into the blacklist.

[0281] In one embodiment, the calculation steps of the risk management model include:

[0282] Based on the warning conditions of the bonds, when the bonds are determined to be warning bonds, the warning level of the warning bonds is obtained;

[0283] When the pension product is of standard type, the warning level of the warning bond is used to determine whether the pension product is a high-risk warning pension product;

[0284] When a pension product is non-standard, when the net value of the pension product is adjusted downward or the daily rate of return is negative, the pension product is determined to be a high-risk warning pension product;

[0285] After determining that the pension product is determined as a high-risk warning pension product, query the associated warning pension products of the pension product to determine that the associated warning pension product is a high-risk warning pension product.

[0286] In one embodiment, determining the bond warning level based on the bond warning condition includes:

[0287] Calculate the value of each type of indicator used for early warning of bonds;

[0288] When the value of the indicator meets the warning condition, the indicator is determined to be a warning indicator, the warning level of the warning indicator is determined, and the bond is determined to be a warning bond;

[0289] The warning level of the warning bond is determined based on the warning levels of all warning indicators of the warning bond.

[0290] In one embodiment, judging whether the pension product is a high-risk warning pension product according to the warning level of the warning bond includes:

[0291] Calculate the excess rate of return of standard pension products;

[0292] Determine the yield of the early warning bonds according to their warning levels;

[0293] Determine the warning triggering conditions based on the yield of the warning bond and the benchmark yield;

[0294] When the excess rate of return of a standard pension product meets the warning triggering condition, it is determined whether the pension product is a high-risk warning pension product.

[0295] In one embodiment, the core pool determination module is specifically used to:

[0296] The proportion of investment managers with optimistic, neutral and pessimistic views in the market opinion data of investment managers corresponding to pension products with each strategy style label is determined as the proportion of pension products with this strategy style label under different attribute labels.

[0297] In one embodiment, the core pool determination module is specifically used to:

[0298] The proportion of investment managers with different operating styles expected in the investment manager market opinion data corresponding to the pension products with each strategy style label is determined as the proportion of the pension products with this strategy style label.

[0299] In one embodiment, the core pool determination module is specifically used to:

[0300] Calculate the risk-return index value of each pension product in the candidate pool;

[0301] The pension products selected into the core pool are determined based on the proportion of pension products with each strategy style label under different attribute labels, as well as the risk-return indicator value of pension products with each strategy style label.

[0302] Fig.11 Schematic diagram of a pension product screening device in an embodiment of the present invention Figure 3 In one embodiment, the device further comprises a display module 1101, which is used to:

[0303] Displays the strategy style labels and attribute labels, alternative pools and core pools of pension products.

[0304] In summary, in the device proposed in the embodiment of the present invention, the investment data of pension products are obtained; based on the investment data, the strategy style labels of pension products are analyzed; the attribute labels of pension products with different strategy style labels are analyzed; according to the whitelist access requirements corresponding to different strategy style labels, the pension products that meet the whitelist access requirements are added to the alternative pool; the pension product blacklist is queried, and the pension products in the alternative pool that are on the pension product blacklist are removed; according to the market view data of the investment manager, the proportion of pension products with different strategy style labels under different attribute labels is determined, and the pension products with corresponding proportions in the alternative pool are added to the core pool; and the alternative pool and the core pool are output. In the above process, multiple types of pension products are progressively screened into the alternative pool and the core pool, thereby providing a more accurate performance classification, which is convenient for the subsequent key tracking and analysis of the core pool products.

[0305] An embodiment of the present invention further provides a computer device, Fig.12 It is a schematic diagram of a computer device in an embodiment of the present invention. The computer device 1200 includes a memory 1210, a processor 1220, and a computer program 1230 stored in the memory 1210 and executable on the processor 1220. When the processor 1220 executes the computer program 1230, the above-mentioned pension product screening method is implemented.

[0306] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned pension product screening method is implemented.

[0307] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned pension product screening method is implemented.

[0308] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented 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.

[0309] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0310] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0311] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0312] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A pension product screening method, characterized in that: include: Obtain investment data for pension products; Analyze the strategy style labels of pension products based on the investment data; Analyze the attribute labels of pension products with different strategy style labels; According to the whitelist access requirements corresponding to different strategy style labels, pension products that meet the whitelist access requirements are added to the candidate pool; Check the pension product blacklist and remove the pension products on the pension product blacklist from the candidate pool; According to the market view data of the investment manager, determine the proportion of pension products with different strategy style labels under different attribute labels, and add the corresponding proportion of pension products in the alternative pool to the core pool; Output the candidate pool and core pool; The method further includes: inputting the pension product into the risk management model, determining whether the pension product is a high-risk warning pension product, and placing the high-risk warning pension product into a blacklist; The calculation steps of the risk management model include: based on the early warning conditions of the bond, when the bond is determined to be a warning bond, obtaining the early warning level of the early warning bond; when the pension product is a standard type, judging whether the pension product is a high-risk early warning pension product according to the early warning level of the early warning bond; when the pension product is a non-standard type, when the net value of the pension product is adjusted downward or the daily rate of return is negative, determining that the pension product is a high-risk early warning pension product; after determining that the pension product is determined to be a high-risk early warning pension product, querying the associated early warning pension product of the pension product, and determining that the associated early warning pension product is a high-risk early warning pension product; Standard type refers to pension products with the primary strategy style of absolute return and the secondary strategy style of market value; and pension products with the primary strategy style of relative return and the secondary strategy style of mixed type; Adding the corresponding proportion of pension products in the alternative pool to the core pool, including: calculating the risk-return index value of each pension product in the alternative pool; determining the pension products selected into the core pool according to the proportion of pension products with each strategy style label under different attribute labels and the risk-return index value of pension products with each strategy style label.

2. The pension product screening method according to claim 1, characterized in that: The investment data includes at least one of the following data: Basic information, net value and size data, asset distribution data, heavy holdings data, and market public data.

3. The pension product screening method according to claim 1, characterized in that: Access investment data for pension products, including: The investment data of pension products from different sources are merged, deduplicated and completed.

4. The pension product screening method according to claim 1, characterized in that: The strategy style label includes three levels, among which, The primary strategy styles include absolute return and relative return; The secondary strategy styles under the absolute return type include stock type, mixed type, currency type, market value pricing type, and cost pricing type; the tertiary strategy styles under the market value pricing type include convertible bond type, allocation tool type, and ordinary type; the tertiary strategy styles under the cost pricing type include debt pool type, single debt plan type, trust pool type, single trust type, and deposit type; The secondary strategy styles under the relative return type include stock type and mixed type; the tertiary strategy styles under the stock type include Hong Kong stocks type, specific industry type, index enhancement type, flexible allocation type, value large-cap type, value small-cap type, growth large-cap type, and growth small-cap type; the tertiary strategy styles under the mixed type include allocation tool type, aggressive type, and fixed income + tool type.

5. The pension product screening method according to claim 1, characterized in that: Based on the investment data, analyze the strategy style labels of pension products, including: The investment data is input into a strategy style analysis model to obtain a strategy style label for the pension product.

6. The pension product screening method according to claim 5, characterized in that: The calculation steps of the strategy style analysis model include: Check the registration status of the pension product with the regulatory agency and determine the strategy style label of the pension product; When the strategy style label of the pension product cannot be determined based on the record filing with the regulatory agency, the strategy style label of the pension product is determined according to the calculation method of the strategy style labels of various types of pension products.

7. The pension product screening method according to claim 1, characterized in that: The attribute labels include one or any combination of excess return type, market type, manager performance type, net asset size type, heavily invested industry type, heavily invested stock type, time-weighted annualized rate of return ranking type, time-weighted annualized rate of return ranking stability type, drawdown type, duration type, leverage type, credit type, and turnover rate type.

8. The pension product screening method according to claim 1, characterized in that: Based on the bond's warning conditions, determine the bond's warning level, including: Calculate the value of each type of indicator used for early warning of bonds; When the value of the indicator meets the warning condition, the indicator is determined to be a warning indicator, the warning level of the warning indicator is determined, and the bond is determined to be a warning bond; The warning level of the warning bond is determined based on the warning levels of all warning indicators of the warning bond.

9. The pension product screening method according to claim 1, characterized in that: According to the warning level of the warning bond, determine whether the pension product is a high-risk warning pension product, including: Calculate the excess rate of return of standard pension products; Determine the yield of the early warning bonds according to their warning levels; Determine the warning triggering conditions based on the yield of the warning bond and the benchmark yield; When the excess rate of return of a standard pension product meets the warning triggering condition, it is determined whether the pension product is a high-risk warning pension product.

10. The pension product screening method according to claim 1, characterized in that: Based on the market view data of investment managers, determine the proportion of pension products with different strategy style labels under different attribute labels, including: The proportion of investment managers with optimistic, neutral and pessimistic views in the market opinion data of investment managers corresponding to pension products with each strategy style label is determined as the proportion of pension products with this strategy style label under different attribute labels.

11. The pension product screening method according to claim 1, characterized in that: Based on the market view data of investment managers, determine the proportion of pension products with different strategy style labels under different attribute labels, including: The proportion of investment managers with different operating styles expected in the investment manager market opinion data corresponding to the pension products with each strategy style label is determined as the proportion of the pension products with this strategy style label.

12. The pension product screening method according to claim 1, characterized in that: Also includes: Displays the strategy style labels and attribute labels, alternative pools and core pools of pension products.

13. A pension product screening device, characterized in that: include: Investment data acquisition module, used to obtain investment data of pension products; A strategy style label analysis module, used to analyze the strategy style label of the pension product based on the investment data; Attribute label analysis module, used to analyze the attribute labels of pension products with different strategy style labels; The candidate pool preliminary selection module is used to add pension products that meet the whitelist access requirements to the candidate pool according to the whitelist access requirements corresponding to different strategy style labels; The candidate pool screening module is used to query the pension product blacklist and remove the pension products on the pension product blacklist from the candidate pool; The core pool determination module is used to determine the proportion of pension products with different strategy style labels under different attribute labels based on the market view data of the investment manager, and add the corresponding proportion of pension products in the alternative pool to the core pool; Output module, used for outputting the candidate pool and the core pool; The device also includes a risk management module, which is used to: input the pension product into the risk management model, determine whether the pension product is a high-risk warning pension product, and put the high-risk warning pension product into a blacklist; The calculation steps of the risk management model include: based on the early warning conditions of the bonds, when the bonds are determined to be early warning bonds, obtaining the early warning level of the early warning bonds; When the pension product is of standard type, determine whether the pension product is a high-risk warning pension product based on the warning level of the warning bond; when the pension product is of non-standard type, determine that the pension product is a high-risk warning pension product when the net value of the pension product is adjusted downward or the daily rate of return is negative; after determining that the pension product is determined to be a high-risk warning pension product, query the associated warning pension products of the pension product to determine that the associated warning pension products are high-risk warning pension products; The core pool determination module is specifically used to: calculate the risk-return index value of each pension product in the alternative pool; determine the pension products selected into the core pool based on the proportion of pension products with each strategy style label under different attribute labels and the risk-return index value of pension products with each strategy style label.

14. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 12 is implemented.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 12 is implemented.

16. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 12 is implemented.

Images