WO2020000701A1 - 基金适配方法、系统、计算机设备和存储介质 - Google Patents

基金适配方法、系统、计算机设备和存储介质 Download PDF

Info

Publication number
WO2020000701A1
WO2020000701A1 PCT/CN2018/106590 CN2018106590W WO2020000701A1 WO 2020000701 A1 WO2020000701 A1 WO 2020000701A1 CN 2018106590 W CN2018106590 W CN 2018106590W WO 2020000701 A1 WO2020000701 A1 WO 2020000701A1
Authority
WO
WIPO (PCT)
Prior art keywords
fund
factor
funds
level
score
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2018/106590
Other languages
English (en)
French (fr)
Inventor
蒋逸文
胡逸凡
刘琼
陈泽晖
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2020000701A1 publication Critical patent/WO2020000701A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/06—Asset management; Financial planning or analysis
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange

Definitions

  • the present application relates to the field of financial information technology, and in particular, to a fund adaptation method, system, computer equipment, and storage medium.
  • Quantitative investment refers to a transaction method that uses quantitative methods and computer-programmed orders to buy and sell in order to obtain stable returns. All decisions to quantify investments are based on models. Its models generally include major asset allocation models, industry models, and stock models.
  • the basic logic of the model is: determine how many funds to choose for each index based on the investment amount, and decide which one to choose based on the fund's ranking in the index. fund.
  • rankings for fund screening models are generally provided through investment pools for screening.
  • a single ranking strategy is usually used. For example, after ranking only the annualized rate or volatility, the top ranked funds are selected to be placed in the investment pool. This ranking strategy is more one-sided, It is easy to cause risks in subsequent asset allocation.
  • a fund adaptation method includes:
  • a configuration file is set in the financial transaction system, and the configuration file contains first-level factors and ranking rules for ranking each type of fund, at least one second-level factor is set for each first-level factor, and each second-level factor is set Score calculation script;
  • the fund calculates equal weight values in the second factor group, and sums all equal weight values in the fund within the same first factor, to obtain the first factor score of the fund;
  • All funds in the same type of funds are ranked according to the total score of the funds, and a predetermined number of funds with a higher ranking are selected and put into an investment pool for adaptation.
  • a fund adaptation system includes:
  • the configuration unit is configured to set a configuration file in the financial transaction system.
  • the configuration file contains a first-level factor and a ranking rule for ranking each type of fund, and at least one second-level factor is set for each first-level factor.
  • a second-level factor score calculation unit configured to call the score calculation script to perform a second-level factor score calculation on all individual funds of a type of fund in the fund trading system one by one to obtain a second-level factor score of the fund;
  • the first-level factor score calculation unit is configured to read the ranking rule, use the ranking rule to rank the second-level factor score, and filter out a preset number of funds with a higher ranking of the second-level factor score to be set to two.
  • Grade factor group the equal weight values of the selected funds in the second factor group are calculated, and all equal weight values in the same first factor of the fund are summed to obtain a value of the fund.
  • Grade factor score the equal weight values of the selected funds in the second factor group are calculated, and all equal weight values in the same first factor of the fund are summed to obtain a value of the fund.
  • a total score calculation unit configured to normalize each of the first-level factor scores of the fund, and then sum up each of the first-level factor scores to obtain a total score of the fund;
  • the screening unit is configured to rank all funds in the same type of fund according to the total score of the funds from high to low, and select a preset number of funds in the top ranking into the investment pool for adaptation.
  • a computer device includes a memory and a processor.
  • the memory stores computer-readable instructions.
  • the processor causes the processor to perform the following steps:
  • a configuration file is set in the financial transaction system, and the configuration file contains first-level factors and ranking rules for ranking each type of fund, at least one second-level factor is set for each first-level factor, and each second-level factor is set Score calculation script;
  • the fund calculates equal weight values in the second factor group, and sums all equal weight values in the fund within the same first factor, to obtain the first factor score of the fund;
  • All funds in the same type of funds are ranked according to the total score of the funds, and a predetermined number of funds with a higher ranking are selected and put into an investment pool for adaptation.
  • a storage medium storing computer-readable instructions.
  • the one or more processors execute the following steps:
  • a configuration file is set in the financial transaction system, and the configuration file contains first-level factors and ranking rules for ranking each type of fund, at least one second-level factor is set for each first-level factor, and each second-level factor is set Score calculation script;
  • the fund calculates equal weight values in the second factor group, and sums all equal weight values in the fund within the same first factor, to obtain the first factor score of the fund;
  • All funds in the same type of funds are ranked according to the total score of the funds from high to low, and a predetermined number of funds with a higher ranking are selected and placed in the investment pool for adaptation.
  • the above-mentioned fund adaptation method, device, computer equipment, and storage medium include setting a configuration file in a financial transaction system.
  • the configuration file contains first-level factors and ranking rules for ranking each type of fund, and at least each first-level factor is set.
  • a second-level factor setting a score calculation script for each second-level factor; invoking the score calculation script to perform a second-level factor score calculation on all individual funds of a type of fund in the fund trading system one by one to obtain the second-level factor score of the fund; Read the ranking rules, use the ranking rules to rank the secondary factor scores, and select a preset number of funds with a higher ranking of the secondary factor scores as the secondary factor group, and place the selected funds in the secondary factor group.
  • FIG. 1 is a flowchart of a fund adaptation method according to an embodiment of the present application
  • FIG. 2 is a structural diagram of a fund adaptation system in an embodiment of the present application.
  • FIG. 1 is a flowchart of a fund adaptation method according to an embodiment of the present application. As shown in FIG. 1, a fund adaptation method includes the following steps:
  • Step S1 a configuration file: a configuration file is set in the financial transaction system, the configuration file contains a first-level factor and a ranking rule for ranking each type of fund, at least one second-level factor is set for each first-level factor, and each Second factor setting score calculation script.
  • the fund transaction system in this step may be any system in the prior art that can perform multiple types of funds, such as currency, equity, and bond fund transactions.
  • a configuration file is set in the database in advance to configure all individual funds that can be bought and sold in the fund trading system, and first-level factors, ranking rules, second-level factors, and score calculation scripts are preset.
  • the configuration file for this step can be as shown in Table 1 below:
  • Scoring calculation script 1 Annualized rate of return (nearly one year / half year / three months).
  • the annualized rate of return refers to converting the cumulative income of a single fund into an annual rate of return.
  • Scoring calculation script 2 Cumulative yield (nearly one year / six months / three months). Cumulative yield refers to the cumulative income of a single fund in the past one year / six months / three months to date, including changes in cash dividend returns and fund net worth The cumulative yield can be a measure of the fund's returns since its establishment.
  • Scoring calculation script 3 Sharpe ratio (near one / two / three years), this ratio is to calculate how much excess return can be generated for each unit of total portfolio risk.
  • Scoring calculation script 4 Sotino ratio (near one / two / three years), similar to the Sharpe ratio, but this ratio is to calculate how much excess return can be generated for each unit of the portfolio that bears less than risk-free returns.
  • Scoring calculation script 5 Calmar ratio (near one / two / three years).
  • the Kalmar ratio describes the relationship between the return and the maximum retracement.
  • the calculation method is the annualized return and the historical maximum retracement. Ratio.
  • Scoring calculation script 6 Jensen index (nearly one / two / three years).
  • the Jensen index is the excess return on fund performance that exceeds market benchmark portfolios.
  • Scoring Script 7 Maximum retracement (near one / two / three years), describing the worst possible scenario for the strategy, the most extreme possible loss scenario.
  • Scoring calculation script 8 Volatility (near one / two years), that is, the standard deviation, reflects the fluctuation range of the total return during the calculation period, that is, the degree of deviation of the total monthly return of the fund from the average monthly return.
  • Scoring calculation script 9 Downward volatility (close to one year), that is, downside risk. When calculating the standard deviation, the mean value is not used, and the part exceeding the acceptable minimum return rate is excluded.
  • Scoring calculation script 10 fund size, fund size.
  • Scoring calculation script 11 the duration of the fund, the length of time that the fund lasts until the ranking.
  • Scoring calculation script 12 Time of incumbent fund manager and length of time of incumbent fund manager.
  • Score calculation script 13 Fund manager's timing ability, call T-M model, fund manager's ability to predict the difference between market returns and risk-free returns.
  • Scoring calculation script 14 The fund manager's ability to select stocks and call the T-M model. The difference in the return of the portfolio with equal fund returns and system risk. The larger the value, the stronger the stock selection ability.
  • Score calculation script 15 Morningstar's three-year rating, calling the Morningstar rating system to obtain the ranking of the risk-adjusted return of a single fund in its category.
  • Step S2 Calculate the second-level factor score: call the score calculation script to calculate the second-level factor score for all individual funds of a type of fund in the fund trading system one by one to obtain the second-level factor score of the fund.
  • Fund types are divided into currency, equity, and bonds. This step ranks the funds of the same type, calculates the secondary factors of all individual funds in such funds, and calculates all secondary factors of each fund. Score.
  • Step S3 Calculate the first-level factor score: read the ranking rules, use the ranking rules to rank the second-level factor scores, and filter out a preset number of funds with the second-level factor scores as the second-level factor group.
  • the output funds are equal weighted in the second-level factor group, and all the equal weighted values within the same first-level factor in the fund are summed to obtain the first-level factor score of the fund.
  • the ranking rules of each second-level factor are different, some are ranked from high to low, and some are ranked from low to high. Therefore, in this step, the calculated second-level factor score is calculated by reading the ranking rules in the configuration file. Ranking, screening out a preset number of funds as a secondary factor group. The preset number can be determined according to the amount of such funds in the fund trading system, such as 10, 30, or 50.
  • Equal weight means that the constituent members in an index share the same weight, and the equal weight value is the weighted result obtained.
  • the equal weight value of the preset number of funds selected depends on the value of the preset number. For example, when the preset number is 10, it means that there are 10 funds in the secondary factor group, then The equal weight value of each fund is 1/10, that is, the equal weight value is 0.1.
  • the preset number is two.
  • the ranking rules for annualized returns are from high to low.
  • the annualized returns are used to select funds 1 and 3 as annualized returns.
  • Group, Fund 1 and Fund 3 have equal weight values in the annualized return group of 0.5. Calculate the equal weight values corresponding to the second-level factors in Table 1, as shown in Table 3 below:
  • the return factor score of fund 1 is 1.5 and the risk factor score is 0; the return factor score of fund 2 is 0.5 and the risk factor score is 1; the return factor score of fund 3 is 1 and the risk factor score is 0; the return of fund 4 is The factor score is 0 and the risk factor score is 1.
  • Step S4 Calculate the total score: after normalizing the scores of each first-level factor of the fund, sum up the scores of each first-level factor to obtain the total score of the fund.
  • Normalization is a way to simplify calculations. That is, a dimensional expression is transformed into a dimensionless one. The expression becomes scalar, and the normalized value is between 0-1. In this step, the normalized score of each first-level factor is summed to obtain a more accurate total score.
  • the value of the return factor of Fund 1 is 0.6, and the value of the risk factor is 0; the value of the return factor of Fund 2 is 0.2, and the value of the risk factor is 0.4; Fund 3 The value of the return factor is 0.4 and the value of the risk factor is 0; the value of the return factor of fund 4 is 0 and the value of the risk factor is 0.4.
  • step S5 the funds are selected: all funds in the same type of funds are ranked according to the total score of the funds, and a predetermined number of funds with a higher ranking are selected and put into an investment pool for adaptation.
  • the value of the preset number in this step may be the same as or different from the value of the preset number in step S3.
  • the value of the preset number in this step can be determined according to the size of the selection amount for adaptation. For example, taking 4 funds in Table 1 as an example, in this step, one fund can be preset to be placed in the investment pool. For example, when there are several points with the highest total score, the highest ranked ones are put into the investment pool.
  • step S4 the total scores of the four funds are obtained. After ranking from high to low, the total scores of fund 1 and fund 2 are both 0.6, so both fund 1 and fund 2 can be put into the investment pool for adaptation.
  • a multiple-fitting strategy is performed on each of these funds, and after obtaining standardized scores, the total score of each fund is calculated, and the top several funds of this type are selected.
  • Funds are used as investment pools for screening during asset allocation.
  • the first-level factor includes at least one of a return factor or a risk factor
  • the second-level factor of the return factor includes an annualized return rate, a cumulative return rate, a Sharpe ratio, a Sotino ratio, a Kalmar ratio, or a Zhan
  • the ranking rule of the return factor is to rank according to the score from high to low
  • the second-level factor of the risk factor includes at least one of the maximum retracement rate, volatility or downward volatility, risk factor
  • the ranking rules are ranked from low to high score.
  • This embodiment lists the main first-level factors including return factors and risk factors, and defines the ranking rules of the two factors. As shown in Table 1, in addition to the above two factors, this application may include other first-level factors and second-level factors. Order factor. Specifically, various different primary factors and corresponding secondary factors can be selected according to requirements.
  • risk and return are the two most important factors. From the perspective of asset classes, the risk and return of financial products are directly proportional, that is, the greater the risk, the greater the return. But this does not mean that risks and returns must be directly proportional, there are some high-yield but low-risk investment opportunities.
  • several high-return and low-risk funds are selected as investment pools for screening during asset allocation, which can effectively avoid subsequent risk problems.
  • a configuration interface is preset in the fund trading system, a configuration file uploaded to the fund trading system is obtained through the configuration interface, and the configuration file is stored in a database of the fund trading system.
  • the configuration file can be used as shown in Table 1.
  • the configuration interface can include an upload interface for uploading a configuration file.
  • the configuration file is uploaded to the configuration interface through the upload interface. After that, it is stored in the database for use.
  • the configuration interface can also be added to modify and delete field information of the first-level factor, second-level factor, and ranking rules in the configuration file.
  • the preset configuration interface can implement the functions of pre-storing, modifying, and deleting configuration files.
  • a quarter is used as a time period, and after the score calculation script is periodically called, a preset number of funds with a high ranking are selected and put into an investment pool for adaptation.
  • each quarter is a time period and the total score of each fund is calculated regularly. Rankings are selected, and a number of funds are regularly selected into the investment pool for adaptation on a quarterly basis.
  • N is the number of funds in the second factor group.
  • Equal weight value is an equal weight index compilation method, which refers to equal results of the constituent members in an index.
  • market value-weighted index compilation methods are generally adopted, that is, the market value of listed companies is used to calculate weights.
  • the equal weight index method avoids the phenomenon that the larger the component stock market value, the higher the weight in the market value weighting method.
  • the market value of the fund is equal to the share capital times the fund price, that is, the higher the fund price, the greater the weight of the fund in the index. This goes against the principle of value investment.
  • the performance of small and medium-cap companies is often better than that of large-cap companies. Therefore, in a certain period, the performance of the equal weight index can also exceed the market value weight index.
  • This embodiment adopts the above-mentioned equal-weight method, and adjusts the number of funds in the secondary factor group to treat the funds with different market values in the same manner, avoiding that the funds selected are all for the purpose of high prices, thereby avoiding investment risks.
  • a weight calculation is also performed: the first-level factor score of the fund is normalized After the normalization process, read the weight value corresponding to the first-level factor, multiply the value obtained by the normalization process by the weight value, and then sum the scores of each first-level factor to get the total score of the fund. Set in the configuration file.
  • Weight refers to the degree of importance of a first-level factor relative to the investment pool. It is different from the general weight and reflects not only the percentage of a first-level factor. It emphasizes the relative importance of the first-level factor. On contribution or importance.
  • a weight value corresponding to each first-level factor is also preset in a configuration file stored in a database in advance.
  • the weight value of the return factor can be a value
  • the weight value of the risk factor can be b value.
  • step S4 after normalizing the scores of each of the first-level factors of the fund, before summing the scores of each of the first-level factors, multiply the value obtained by the normalization processing by the weight value, and at this time, get The results are as follows: the value of the return factor of Fund 1 is 0.6 ⁇ a and the value of the risk factor is 0; the value of the return factor of Fund 2 is 0.26 ⁇ a and the value of the risk factor is 0.4 ⁇ b; , The value of the risk factor is 0; the value of the return factor of fund 4 is 0, and the value of the risk factor is 0.4 ⁇ b.
  • the value obtained by normalizing the value multiplied by the weight value and then summing the scores of each first-level factor to obtain the total score of the fund can highlight the importance of some first-level factors to the investment pool. For example, for a risk-type investment pool, the weight value of the risk factor is relatively high. For example, for a stable investment pool, the weight value of the risk factor is relatively low. With the above design, different types of investment pools can be obtained for adaptation.
  • a fund adaptation system is proposed, as shown in FIG. 2, and includes the following units:
  • the configuration unit is set to set a configuration file in the financial trading system.
  • the configuration file contains first-level factors and ranking rules for ranking each type of fund, at least one second-level factor is set for each first-level factor, and two for each item.
  • Grade factor setting score calculation script ;
  • the second factor score calculation unit is set to call a score calculation script to calculate the second factor scores of all the individual funds of a type of fund in the fund trading system one by one to obtain the second factor scores of the funds;
  • the first-level factor score calculation unit is set to read the ranking rules, use the ranking rules to rank the second-level factor scores, and filter out the preset number of funds with the second-level factor scores in the top-ranked set to be the second-level factor groups.
  • the out-of-funds find equal-weighted values in the second-level factor group, and sum up all the equal-weighted values in the same first-level factor of the fund to obtain the first-level factor score of the fund;
  • the total score calculation unit is set to normalize the scores of each first-level factor of the fund, and then sum up the scores of each first-level factor to obtain the total score of the fund;
  • the screening unit is configured to rank all funds in the same type of fund according to the total score of the funds from high to low, and select a preset number of funds in the top ranking into the investment pool for adaptation.
  • an equal weight value calculation unit is further included:
  • the equal weight value calculation unit is set to calculate the equal weight value of the selected funds in the secondary factor group, using the following calculation formula:
  • N is the number of funds in the secondary factor group.
  • the first-level factor includes at least one of a return factor or a risk factor
  • the second-level factor of the return factor includes an annualized return rate, a cumulative return rate, a Sharpe ratio, a Sotino ratio, a Kalmar ratio, or a Zhan
  • the ranking rule of the return factor is to rank according to the score from high to low
  • the second-level factor of the risk factor includes at least one of the maximum retracement rate, volatility or downward volatility, risk factor
  • the ranking rules are ranked from low to high score.
  • the configuration unit is further configured to preset a configuration interface in the fund trading system, obtain a configuration file uploaded to the fund trading system through the configuration interface, and store the configuration file in a database of the fund trading system.
  • the second-level factor score calculation unit is further configured to periodically call the score calculation script on a quarterly basis as a time period, and then screen out a preset number of funds that are ranked in the top of the investment pool for adaptation.
  • it further includes a weight calculation unit, configured to normalize the scores of each first-level factor of the fund, read the weight value corresponding to the first-level factor, and multiply the value obtained by the normalization process. Take the weight value and sum up the scores of each first-level factor to get the total score of the fund.
  • the weight value is preset in the configuration file.
  • a computer device which includes a memory and a processor.
  • the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor is caused to implement the foregoing when the computer-readable instructions are executed. Steps in the fund adaptation method in the embodiment.
  • a storage medium storing computer-readable instructions.
  • the computer-readable instructions are executed by one or more processors, the one or more processors are caused to execute the fund adaptation in the foregoing embodiments. Steps in the method.
  • the storage medium may be a non-volatile storage medium.
  • the program may be stored in a computer-readable storage medium.
  • the storage medium may include: Read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks, etc.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Technology Law (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Human Resources & Organizations (AREA)
  • Operations Research (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

金融信息技术领域的一种基金适配方法、系统、计算机设备和存储介质。基金适配方法包括：对一类基金的基金逐个进行二级因子得分计算，得到基金的二级因子得分；对二级因子得分进行排名，筛选出预设个数的基金，将筛选出的基金求等权重值，将基金中的所有等权重值进行求和，得到基金的一级因子得分；将基金的每项一级因子得分进行归一化处理后，对每项一级因子得分进行求和，得到基金的总得分；按基金的总得分从高到低进行排名，筛选出预设个数的基金放入投资池内供适配。通过对一类基金中的每个基金都计算得分，优化投资池内的供选择基金，可大大避免资产配置时的风险，实现投资收益最大化。

Description

基金适配方法、系统、计算机设备和存储介质
本申请要求于2018年06月25日提交中国专利局、申请号为201810656996.X、发明名称为“基金适配方法、系统、计算机设备和存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及金融信息技术领域，尤其涉及一种基金适配方法、系统、计算机设备和存储介质。
背景技术
量化投资是指通过数量化方式及计算机程序化发出买卖指令，以获取稳定收益为目的的交易方式。量化投资的所有决策都是依据模型做出的。其模型一般包括大类资产配置模型、行业模型、股票模型。
在大类资产配置模型中，需要通过基金筛选模型对投资进行监控及调整，该模型的基本逻辑是：根据投资金额决定每类指数要选多少个基金，根据基金在指数当中的排名决定选哪个基金。现有技术中，一般通过投资池为基金筛选模型提供排名，供其进行筛选。但是，投资池内的基金，在排名时，通常采用单一的排名策略，如只对年化率或波动率进行排名后，选取排名靠前的若干基金放入投资池，这种排名策略较为片面，容易造成后续资产配置时的风险。
发明内容
有鉴于此，有必要针对投资池排名策略较为单一，容易出现后续资产配置时的风险问题，提供一种基金适配方法、系统、计算机设备和存储介质。
一种基金适配方法，包括：
在金融交易系统中设置配置文件，所述配置文件内含有对每类基金进行排名的一级因子和排名规则，对每项一级因子设置至少一项二级因子，对每项二级因子设置得分计算脚本；
调用所述得分计算脚本，对所述基金交易系统中一类基金的所有单个基金逐个进行二级因子得分计算，得到所述基金的二级因子得分；
读取所述排名规则，以所述排名规则对二级因子得分进行排名，筛选出所述二级因子得分排名靠前的预设个数的基金设置为二级因子组，将筛选出的所述基金在所述二级因子组内求等权重值，将所述基金中处于同一所述一级因子内的所有等权重值进行求和，得到所述基金的一级因子得分；
将所述基金的每项所述一级因子得分进行归一化处理后，对每项所述一级因子得分进行求和，得到所述基金的总得分；
将同一类基金中的所有基金按所述基金的总得分从高到低进行排名，筛选出排名靠前的预设个数的基金放入投资池内供适配。
一种基金适配系统，包括：
配置单元，设置为在金融交易系统中设置配置文件，所述配置文件内含有对每类基金进行排名的一级因子和排名规则，对每项一级因子设置至少一项二级因子，对每项二级因子设置得分计算脚本；
二级因子得分计算单元，设置为调用所述得分计算脚本，对所述基金交易系统中一类基金的所有单个基金逐个进行二级因子得分计算，得到所述基金的二级因子得分；
一级因子得分计算单元，设置为读取所述排名规则，以所述排名规则对二级因子得分进行排名，筛选出所述二级因子得分排名靠前的预设个数的基金设置为二级因子组，将筛选出的所述基金在所述二级因子组内求等权重值，将所述基金的同一项一级因子内的所有等权重值进行求和，得到所述基金的一级因子得分；
总得分计算单元，设置为将所述基金的每项所述一级因子得分进行归一化处理后，对每项所述一级因子得分进行求和，得到所述基金的总得分；
筛选单元，设置为将同一类基金中的所有基金按所述基金的总得分从高到低进行排名，筛选出排名靠前的预设个数的基金放入投资池内供适配。
一种计算机设备，包括存储器和处理器，所述存储器中存储有计算机可读指令，所述计算机可读指令被所述处理器执行时，使得所述处理器执行以下步 骤：
在金融交易系统中设置配置文件，所述配置文件内含有对每类基金进行排名的一级因子和排名规则，对每项一级因子设置至少一项二级因子，对每项二级因子设置得分计算脚本；
调用所述得分计算脚本，对所述基金交易系统中一类基金的所有单个基金逐个进行二级因子得分计算，得到所述基金的二级因子得分；
读取所述排名规则，以所述排名规则对二级因子得分进行排名，筛选出所述二级因子得分排名靠前的预设个数的基金设置为二级因子组，将筛选出的所述基金在所述二级因子组内求等权重值，将所述基金中处于同一所述一级因子内的所有等权重值进行求和，得到所述基金的一级因子得分；
将所述基金的每项所述一级因子得分进行归一化处理后，对每项所述一级因子得分进行求和，得到所述基金的总得分；
将同一类基金中的所有基金按所述基金的总得分从高到低进行排名，筛选出排名靠前的预设个数的基金放入投资池内供适配。
一种存储有计算机可读指令的存储介质，所述计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行以下步骤：
在金融交易系统中设置配置文件，所述配置文件内含有对每类基金进行排名的一级因子和排名规则，对每项一级因子设置至少一项二级因子，对每项二级因子设置得分计算脚本；
调用所述得分计算脚本，对所述基金交易系统中一类基金的所有单个基金逐个进行二级因子得分计算，得到所述基金的二级因子得分；
读取所述排名规则，以所述排名规则对二级因子得分进行排名，筛选出所述二级因子得分排名靠前的预设个数的基金设置为二级因子组，将筛选出的所述基金在所述二级因子组内求等权重值，将所述基金中处于同一所述一级因子内的所有等权重值进行求和，得到所述基金的一级因子得分；
将所述基金的每项所述一级因子得分进行归一化处理后，对每项所述一级因子得分进行求和，得到所述基金的总得分；
将同一类基金中的所有基金按所述基金的总得分从高到低进行排名，筛选 出排名靠前的预设个数的基金放入投资池内供适配。
上述基金适配方法、装置、计算机设备和存储介质，包括在金融交易系统中设置配置文件，配置文件内含有对每类基金进行排名的一级因子和排名规则，对每项一级因子设置至少一项二级因子，对每项二级因子设置得分计算脚本；调用得分计算脚本，对基金交易系统中一类基金的所有单个基金逐个进行二级因子得分计算，得到基金的二级因子得分；读取排名规则，以排名规则对二级因子得分进行排名，筛选出二级因子得分排名靠前的预设个数的基金设置为二级因子组，将筛选出的基金在二级因子组内求等权重值，将基金中处于同一一级因子内的所有等权重值进行求和，得到基金的一级因子得分；将基金的每项一级因子得分进行归一化处理后，对每项一级因子得分进行求和，得到基金的总得分；将同一类基金中的所有基金按基金的总得分从高到低进行排名，筛选出排名靠前的预设个数的基金放入投资池内供适配。本技术方案基于多因子拟合策略，通过对一类基金中的每个基金都计算得分，优化投资池内的供选择基金，可大大避免资产配置时的风险，实现投资收益最大化。
附图说明
通过阅读下文优选实施方式的详细描述，各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。附图仅用于示出优选实施方式的目的，而并不认为是对本申请的限制。
图1为本申请一个实施例中的基金适配方法的流程图；
图2为本申请一个实施例中基金适配系统的结构图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
本技术领域技术人员可以理解，除非特意声明，这里使用的单数形式“一”、“一个”、“所述”和“该”也可包括复数形式。应该进一步理解的是，本申请 的说明书中使用的措辞“包括”是指存在所述特征、整数、步骤、操作、元件和/或组件，但是并不排除存在或添加一个或多个其他特征、整数、步骤、操作、元件、组件和/或它们的组。
图1为本申请一个实施例中的基金适配方法的流程图，如图1所示，一种基金适配方法，包括以下步骤：
步骤S1，配置文件：在金融交易系统中设置配置文件，配置文件内含有对每类基金进行排名的一级因子和排名规则，对每项一级因子设置至少一项二级因子，对每项二级因子设置得分计算脚本。
本步骤中的基金交易系统可以是任一现有技术中的可以进行多种类型的基金，如货币类、权益类、债券类基金交易的系统。在基金交易系统中，预先在数据库中设置配置文件，对基金交易系统中能买卖的所有单个基金进行配置，分别预设一级因子、排名规则、二级因子及得分计算脚本。本步骤的配置文件可以采用如下表1所示：
Figure PCTCN2018106590-appb-000001
Figure PCTCN2018106590-appb-000002
表1
其中，对二级因子配置得分计算脚本时，根据如下解释配置：
得分计算脚本1：年化收益率(近一年/半年/三个月)，年化收益率是指将单个基金的累计收益折算为每年的收益率。
得分计算脚本2：累计收益率(近一年/半年/三个月)，累计收益率是指单个基金近一年/半年/三个月至今的累计收益，其中包括现金分红收益、基金净值变化产生的收益，累计收益率可以衡量基金从成立以来的收益情况。
得分计算脚本3：夏普比率(近一/两/三年)，该比率是计算投资组合每承受一单位总风险，能够产生多少的超额报酬。
得分计算脚本4：索提诺比率(近一/两/三年)，与夏普比率类似，但该比率是计算投资组合每承受一单位低于无风险收益的风险，能够产生多少超额报酬。
得分计算脚本5：卡尔玛(Calmar)比率(近一/两/三年)，卡尔玛比率描述的是收益和最大回撤之间的关系，计算方式为年化收益率与历史最大回撤之间的比率。
得分计算脚本6：詹森指数(近一/两/三年)，詹森指数是基金业绩中超过市场基准组合所获得的超额收益。
得分计算脚本7：最大回撤(近一/两/三年)，描述策略可能出现的最糟糕的情况，最极端可能的亏损情况。
得分计算脚本8：波动率(近一/两年)，即标准差，反映计算期内总回报率的波动幅度，即基金每月的总回报率相对于平均月回报率的偏差程度。
得分计算脚本9：下行波动率(近一年)，即下行风险，计算标准差的时候，不采用均值，剔除超出可接受最小收益率的部分。
得分计算脚本10：基金规模，基金的规模大小。
得分计算脚本11：基金存续时间，基金成立到排名时持续的时间长度。
得分计算脚本12：在任基金经理任职时间，在任基金经理的任职时长。
得分计算脚本13：基金经理择时能力，调用T-M模型，基金经理预测市场收益与无风险收益之间差异大小的能力。
得分计算脚本14：基金经理择股能力，调用T-M模型，基金收益与系统风险相等的投资组合收益率的差异，值越大，选股能力越强。
得分计算脚本15：晨星三年评级，调用晨星评级系统，获取单个基金的风险调整后收益在其所在分类下的排名。
步骤S2，计算二级因子得分：调用得分计算脚本，对基金交易系统中一类基金的所有单个基金逐个进行二级因子得分计算，得到基金的二级因子得分。
基金类型分为货币类、权益类、债券类等，本步骤以同一类基金进行排名，将此类基金中的所有单个基金的二级因子均进行计算，计算出每个基金的所有二级因子得分。
具体的，可以选择如下表2中基金1-基金4，并选择2种一级因子，分别包括收益因子中的二级因子年化收益率、夏普比率、詹森指数，风险因子中的最大回撤率、波动率。通过调用得分计算脚本1、得分计算脚本3、得分计算脚本6、得分计算脚本7、得分计算脚本8，得到所有二级因子得分如下表2所示：
基金名称 年化收益率 夏普比率 詹森指数 最大回撤率 波动率
基金1 168.88％ 2.91 1.59 26.08％ 56.92％
基金2 21.63％ 3.61 0.18 1.76％ 5.15％
基金3 32.18％ 1.30 0.47 15.66％ 22.48％
基金4 25.60％ 1.29 0.28 9.06％ 17.56％
表2
步骤S3，计算一级因子得分：读取排名规则，以排名规则对二级因子得分进行排名，筛选出二级因子得分排名靠前的预设个数的基金设置为二级因子组，将筛选出的基金在二级因子组内求等权重值，将基金中处于同一一级因子内的所有等权重值进行求和，得到基金的一级因子得分。
每个二级因子的排名规则各不相同，有的是从高到低排列，有的是从低到高排列，因此本步骤，通过读取配置文件中的排名规则，对已计算得到的二级因子得分进行排名，筛选出预设个数的基金为二级因子组。预设的个数可以根 据该类基金在基金交易系统中的数量确定，如10、30或50等数值。
等权重是指一个指数中的组成成员所占比重一样，而等权重值即为得到的比重结果。本步骤中，筛选出的预设个数的基金的等权重值，取决于预设个数的数值，比如预设个数为10个时，则说明二级因子组中有10个基金，则每个基金的等权重值为1/10，即等权重值为0.1。
以表2中的4个基金为例，预设个数为2个，年化收益率的排名规则为从高到低排名，则年化收益率筛选出基金1和基金3为年化收益率组，基金1和基金3在年化收益率组内的等权重值均为0.5。依次计算表1中的二级因子对应的等权重值，如下表3所示：
基金名称 年化收益率 夏普比率 詹森指数 最大回撤率 波动率
基金1 0.5 0.5 0.5    
基金2   0.5   0.5 0.5
基金3 0.5   0.5    
基金4       0.5 0.5
表3
则基金1的收益因子得分为1.5，风险因子得分为0；基金2的收益因子得分为0.5，风险因子得分为1；基金3的收益因子得分为1，风险因子得分为0；基金4的收益因子得分为0，风险因子得分为1。
步骤S4，计算总得分：将基金的每项一级因子得分进行归一化处理后，对每项一级因子得分进行求和，得到基金的总得分。
由于每个一级因子得到的一级因子得分有多有少，需要进行归一化处理，归一化是一种简化计算的方式，即将有量纲的表达式，经过变换，化为无量纲的表达式，成为标量，归一化处理后的数值在0-1之间。本步骤通过将每项一级因子得分进行归一化处理再求和，能得到较为准确的总得分。
具体的，步骤S3得到的各一级因子进行归一化处理后，基金1的收益因子数值为0.6，风险因子数值为0；基金2的收益因子数值为0.2，风险因子数值为0.4；基金3的收益因子数值为0.4，风险因子数值为0；基金4的收益因子数值为0，风险因子数值为0.4。最后对基金求和，得到基金1的总得分为0.6， 基金2的总得分为0.6，基金3的总得分为0.4，基金4的总得分为0.4。
步骤S5，筛选基金：将同一类基金中的所有基金按基金的总得分从高到低进行排名，筛选出排名靠前的预设个数的基金放入投资池内供适配。
本步骤中预设个数的数值，可以与步骤S3的预设个数的数值相同，也可以不同。本步骤中预设个数的数值，可以根据供适配时的选择量大小确定数值。比如以表1中4个基金为例，本步骤可以预设1个基金放入投资池，如总分排名最高的有几个时，将排名最高的几个都放入投资池。
具体的，步骤S4得到了四个基金的总得分，从高到低排名后，基金1和基金2总得分均为0.6，因此可以将基金1和基金2均放入投资池内供适配。
本实施例，针对某一类基金，对此类基金中的每个基金进行多因此拟合策略，得到标准化得分后，计算每个基金的总得分，选取此类基金中排名靠前的若干个基金，作为投资池，供资产配置时筛选。上述通过对一类基金中的每个基金都计算得分，优化投资池内的供选择基金，可大大避免资产配置时的风险，实现投资收益最大化。
在一个实施例中，一级因子包括收益因子或风险因子中的至少一项；收益因子的二级因子包括年化收益率、累计收益率、夏普比率、索提诺比率、卡尔玛比率或詹森指数中的至少一项，收益因子的排名规则是按得分的从高到低进行排名；风险因子的二级因子包括最大回撤率、波动率或下行波动率中的至少一项，风险因子的排名规则是按得分的从低到高进行排名。
本实施例列举了主要的一级因子包括收益因子和风险因子，限定了两种因子的排名规则，如表1所示，除了上述两种因子外，本申请还可以包括其他一级因子和二级因子。具体可以根据需要选择各种不同的一级因子和对应的二级因子。
在金融领域，风险和收益是最重要的两个因子，从资产类别的层面上来说，金融产品的风险和收益是成正比的，也就是是说，风险越大，收益越大。但是这并不意味着在风险和收益一定是成正比的，存在着一些高收益但是低风险的投资机会。通过本实施例的两种一级因子的选择，筛选得到若干个高收益低风险的基金作为投资池，供资产配置时筛选，能有效避免后续风险问题。
在一个实施例中，在基金交易系统中预设配置界面，通过配置界面获取上传给基金交易系统的配置文件，并将配置文件存储在基金交易系统的数据库中。
在对一类基金进行筛选前，可以预先设置好配置文件和配置界面，配置文件可以采用如表1所示，配置界面可包含上传配置文件的上传接口，通过上传接口将配置文件上传给配置界面后，存储在数据库中待使用。为了便于修改数据库中的配置文件，在配置界面还可以增设修改、删除配置文件中一级因子、二级因子、排名规则的字段信息。如增设搜索一级因子、二级因子、排名规则的搜索字段、增设修改触发按钮、增设删除触发按钮等人机交互界面，以便于修改删除配置文件中的任何信息。本实施例通过预设配置界面，能实现配置文件的预存储和修改删除功能。
在一个实施例中，调用得分计算脚本时，按一个季度为时间段，定时调用得分计算脚本后，筛选出排名靠前的预设个数的基金放入投资池内供适配。
由于现有的基金交易系统中，基金经理人通常采用季报或年报的时间段披露基金基金季度报告或年度报告，因此本实施例，每一个季度为时间段，定时计算每个基金的总得分，并进行排名筛选，以季度定期选取若干基金放入投资池内供适配。
在一个实施例中，将筛选出的基金在二级因子组内求等权重值时，采用如下计算公式：
Figure PCTCN2018106590-appb-000003
其中，N为二级因子组中基金的个数。
等权重值是一种等权重指数编制方法，指一个指数中的组成成员所占比重一样的相等结果。现有计算中，一般采用多采用市值加权的指数编制方法，即以上市公司市值计算权重。与市值加权编制方法相比，等权重指数编制方法规避了市值加权指数编制方法中，成份股市值越大，权重越高的现象。而基金市值又等于股本乘以基金价格，也就是基金价格越高，该基金在指数中的权重也越大。这有悖于价值投资原则。中小市值公司中长期的表现往往优于大市值公司。因而，在一定时期内，等权重指数的表现也能够超过市值加权指数。
本实施例采用上述等权重的方法，并通过调整二级因子组中基金的个数，对市值不同的基金相同对待，避免筛选出的基金均为高价格的目的，从而规避了投资风险。
在一个实施例中，将基金的每项一级因子得分进行归一化处理后，对每项一级因子得分进行求和前，还进行权重计算：将基金的每项一级因子得分进行归一化处理后，读取一级因子对应的权重值，将归一化处理后得到的数值乘以权重值，再对每项一级因子得分进行求和，得到基金的总得分，权重值预设在配置文件中。
权重是指某一一级因子相对于投资池的重要程度，其不同于一般的比重，体现的不仅仅是某一一级因子所占的百分比，强调的是一级因子的相对重要程度，倾向于贡献度或重要性。在基金交易系统中，预先存储在数据库中的配置文件中，还预设了每个一级因子对应的权重值。比如收益因子的权重值可以是a值，风险因子的权重值可以是b值。则步骤S4，对基金的每项一级因子得分进行归一化处理后，对每项一级因子得分进行求和前，将归一化处理后得到的数值乘以权重值，此时，得到的结果如下：基金1的收益因子数值为0.6×a，风险因子数值为0；基金2的收益因子数值为0.26×a，风险因子数值为0.4×b；基金3的收益因子数值为0.4×a，风险因子数值为0；基金4的收益因子数值为0，风险因子数值为0.4×b。
本实施例，通过归一化处理后得到的数值乘以权重值，再对每项一级因子得分进行求和得到基金的总得分的方式，可以突出某些一级因子对于投资池的重要性，比如风险型的投资池，则风险因子的权重值相对高，比如稳定型的投资池，则风险因子的权重值相对低。上述设计，可以得到不同类型的投资池供适配。
在一个实施例中，提出了一种基金适配系统，如图2所示，包括如下单元：
配置单元，设置为在金融交易系统中设置配置文件，配置文件内含有对每类基金进行排名的一级因子和排名规则，对每项一级因子设置至少一项二级因子，对每项二级因子设置得分计算脚本；
二级因子得分计算单元，设置为调用得分计算脚本，对基金交易系统中一类基金的所有单个基金逐个进行二级因子得分计算，得到基金的二级因子得分；
一级因子得分计算单元，设置为读取排名规则，以排名规则对二级因子得分进行排名，筛选出二级因子得分排名靠前的预设个数的基金设置为二级因子组，将筛选出的基金在二级因子组内求等权重值，将基金的同一项一级因子内的所有等权重值进行求和，得到基金的一级因子得分；
总得分计算单元，设置为将基金的每项一级因子得分进行归一化处理后，对每项一级因子得分进行求和，得到基金的总得分；
筛选单元，设置为将同一类基金中的所有基金按所述基金的总得分从高到低进行排名，筛选出排名靠前的预设个数的基金放入投资池内供适配。
在一个实施例中，还包括等权重值计算单元：
等权重值计算单元，设置为将筛选出的基金在二级因子组内求等权重值时，采用如下计算公式：
Figure PCTCN2018106590-appb-000004
其中，N为所述二级因子组中基金的个数。
在一个实施例中，一级因子包括收益因子或风险因子中的至少一项；收益因子的二级因子包括年化收益率、累计收益率、夏普比率、索提诺比率、卡尔玛比率或詹森指数中的至少一项，收益因子的排名规则是按得分的从高到低进行排名；风险因子的二级因子包括最大回撤率、波动率或下行波动率中的至少一项，风险因子的排名规则是按得分的从低到高进行排名。
在一个实施例中，配置单元，还设置为在基金交易系统中预设配置界面，通过配置界面获取上传给基金交易系统的配置文件，并将配置文件存储在基金交易系统的数据库中。
在一个实施例中，二级因子得分计算单元，还设置为按一个季度为时间段，定时调用得分计算脚本后，筛选出排名靠前的预设个数的基金放入投资池内供适配。
在一个实施例中，还包括权重计算单元，设置为将基金的每项一级因子得 分进行归一化处理后，读取一级因子对应的权重值，将归一化处理后得到的数值乘以权重值，再对每项一级因子得分进行求和，得到基金的总得分，权重值预设在配置文件中。
在一个实施例中，提出了一种计算机设备，包括存储器和处理器，存储器中存储有计算机可读指令，计算机可读指令被处理器执行时，使得处理器执行计算机可读指令时实现上述各实施例里基金适配方法中的步骤。
在一个实施例中，提出了一种存储有计算机可读指令的存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行上述各实施例里基金适配方法中的步骤。其中，存储介质可以为非易失性存储介质。
本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分步骤是可以通过程序来指令相关的硬件来完成，该程序可以存储于一计算机可读存储介质中，存储介质可以包括：只读存储器(ROM，Read Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁盘或光盘等。
以上所述实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请一些示例性实施例，其描述较为具体和详细，但并不能因此而理解为对本申请专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种基金适配方法，包括：
    在金融交易系统中设置配置文件，所述配置文件内含有对每类基金进行排名的一级因子和排名规则，对每项一级因子设置至少一项二级因子，对每项二级因子设置得分计算脚本；
    调用所述得分计算脚本，对所述基金交易系统中一类基金的所有单个基金逐个进行二级因子得分计算，得到所述基金的二级因子得分；
    读取所述排名规则，以所述排名规则对二级因子得分进行排名，筛选出所述二级因子得分排名靠前的预设个数的基金设置为二级因子组，将筛选出的所述基金在所述二级因子组内求等权重值，将所述基金中处于同一所述一级因子内的所有等权重值进行求和，得到所述基金的一级因子得分；
    将所述基金的每项所述一级因子得分进行归一化处理后，对每项所述一级因子得分进行求和，得到所述基金的总得分；
    将同一类基金中的所有基金按所述基金的总得分从高到低进行排名，筛选出排名靠前的预设个数的基金放入投资池内供适配。
  2. 根据权利要求1所述的基金适配方法，其中，所述一级因子包括收益因子或风险因子中的至少一项；
    所述收益因子的二级因子包括年化收益率、累计收益率、夏普比率、索提诺比率、卡尔玛比率或詹森指数中的至少一项，所述收益因子的排名规则是按得分的从高到低进行排名；
    所述风险因子的二级因子包括最大回撤率、波动率或下行波动率中的至少一项，所述风险因子的排名规则是按得分的从低到高进行排名。
  3. 根据权利要求1所述的基金适配方法，其中，所述在金融交易系统中设置配置文件，包括：
    在所述基金交易系统中预设配置界面，通过所述配置界面获取上传给所述基金交易系统的配置文件，并将所述配置文件存储在所述基金交易系统的数据库中。
  4. 根据权利要求1所述的基金适配方法，其中，调用所述得分计算脚本，包括：
    按一个季度为时间段，定时调用所述得分计算脚本后，筛选出所述排名靠前的预设个数的基金放入投资池内供适配。
  5. 根据权利要求1所述的基金适配方法，其中，所述将筛选出的所述基金在所述二级因子组内求等权重值，包括如下计算公式：
    Figure PCTCN2018106590-appb-100001
    其中，N为所述二级因子组中基金的个数。
  6. 根据权利要求1所述的基金适配方法，其中，所述将所述基金的每项所述一级因子得分进行归一化处理后，对每项所述一级因子得分进行求和前，还包括：
    将所述基金的每项所述一级因子得分进行归一化处理后，读取所述一级因子对应的权重值，将归一化处理后得到的数值乘以权重值，再对每项所述一级因子得分进行求和，得到所述基金的总得分，所述权重值预设在所述配置文件中。
  7. 一种基金适配系统，包括：
    配置单元，设置为在金融交易系统中设置配置文件，所述配置文件内含有对每类基金进行排名的一级因子和排名规则，对每项一级因子设置至少一项二级因子，对每项二级因子设置得分计算脚本；
    二级因子得分计算单元，设置为调用所述得分计算脚本，对所述基金交易系统中一类基金的所有单个基金逐个进行二级因子得分计算，得到所述基金的二级因子得分；
    一级因子得分计算单元，设置为读取所述排名规则，以所述排名规则对二级因子得分进行排名，筛选出所述二级因子得分排名靠前的预设个数的基金设置为二级因子组，将筛选出的所述基金在所述二级因子组内求等权重值，将所述基金的同一项一级因子内的所有等权重值进行求和，得到所述基金的一级因子得分；
    总得分计算单元，设置为将所述基金的每项所述一级因子得分进行归一化处理后，对每项所述一级因子得分进行求和，得到所述基金的总得分；
    筛选单元，设置为将同一类基金中的所有基金按所述基金的总得分从高到低进行排名，筛选出排名靠前的预设个数的基金放入投资池内供适配。
  8. 根据权利要求7所述的基金适配系统，其中，还包括等权重值计算单元：
    所述等权重值计算单元，设置为将筛选出的所述基金在所述二级因子组内求等权重值时，采用如下计算公式：
    Figure PCTCN2018106590-appb-100002
    其中，N为所述二级因子组中基金的个数。
  9. 根据权利要求7所述的基金适配系统，其中，所述一级因子包括收益因子或风险因子中的至少一项；
    所述收益因子的二级因子包括年化收益率、累计收益率、夏普比率、索提诺比率、卡尔玛比率或詹森指数中的至少一项，所述收益因子的排名规则是按得分的从高到低进行排名；
    所述风险因子的二级因子包括最大回撤率、波动率或下行波动率中的至少一项，所述风险因子的排名规则是按得分的从低到高进行排名。
  10. 根据权利要求7所述的基金适配系统，其中，所述配置单元，还设置为在所述基金交易系统中预设配置界面，通过所述配置界面获取上传给所述基金交易系统的配置文件，并将所述配置文件存储在所述基金交易系统的数据库中。
  11. 根据权利要求7所述的基金适配系统，其中，所述二级因子得分计算单元，还设置为按一个季度为时间段，定时调用所述得分计算脚本后，筛选出所述排名靠前的预设个数的基金放入投资池内供适配。
  12. 根据权利要求7所述的基金适配系统，其中，还包括权重计算单元，设置为将所述基金的每项所述一级因子得分进行归一化处理后，读取所述一级因子对应的权重值，将归一化处理后得到的数值乘以权重值，再对每项所述一级因子得分进行求和，得到所述基金的总得分，所述权重值预设在所述配置文件中。
  13. 一种计算机设备，包括存储器和处理器，所述存储器中存储有计算机 可读指令，所述计算机可读指令被所述处理器执行时，使得所述处理器执行以下步骤：
    在金融交易系统中设置配置文件，所述配置文件内含有对每类基金进行排名的一级因子和排名规则，对每项一级因子设置至少一项二级因子，对每项二级因子设置得分计算脚本；
    调用所述得分计算脚本，对所述基金交易系统中一类基金的所有单个基金逐个进行二级因子得分计算，得到所述基金的二级因子得分；
    读取所述排名规则，以所述排名规则对二级因子得分进行排名，筛选出所述二级因子得分排名靠前的预设个数的基金设置为二级因子组，将筛选出的所述基金在所述二级因子组内求等权重值，将所述基金中处于同一所述一级因子内的所有等权重值进行求和，得到所述基金的一级因子得分；
    将所述基金的每项所述一级因子得分进行归一化处理后，对每项所述一级因子得分进行求和，得到所述基金的总得分；
    将同一类基金中的所有基金按所述基金的总得分从高到低进行排名，筛选出排名靠前的预设个数的基金放入投资池内供适配。
  14. 根据权利要求13所述的计算机设备，其中，所述在金融交易系统中设置配置文件时，使得所述处理器执行以下步骤：
    在所述基金交易系统中预设配置界面，通过所述配置界面获取上传给所述基金交易系统的配置文件，并将所述配置文件存储在所述基金交易系统的数据库中。
  15. 根据权利要求13所述的计算机设备，其中，所述调用所述得分计算脚本时，使得所述处理器执行以下步骤：
    按一个季度为时间段，定时调用所述得分计算脚本后，筛选出所述排名靠前的预设个数的基金放入投资池内供适配。
  16. 根据权利要求13所述的计算机设备，其中，所述将筛选出的所述基金在所述二级因子组内求等权重值时，使得所述处理器执行如下计算公式：
    Figure PCTCN2018106590-appb-100003
    其中，N为所述二级因子组中基金的个数。
  17. 一种存储有计算机可读指令的存储介质，所述计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行以下步骤：
    在金融交易系统中设置配置文件，所述配置文件内含有对每类基金进行排名的一级因子和排名规则，对每项一级因子设置至少一项二级因子，对每项二级因子设置得分计算脚本；
    调用所述得分计算脚本，对所述基金交易系统中一类基金的所有单个基金逐个进行二级因子得分计算，得到所述基金的二级因子得分；
    读取所述排名规则，以所述排名规则对二级因子得分进行排名，筛选出所述二级因子得分排名靠前的预设个数的基金设置为二级因子组，将筛选出的所述基金在所述二级因子组内求等权重值，将所述基金中处于同一所述一级因子内的所有等权重值进行求和，得到所述基金的一级因子得分；
    将所述基金的每项所述一级因子得分进行归一化处理后，对每项所述一级因子得分进行求和，得到所述基金的总得分；
    将同一类基金中的所有基金按所述基金的总得分从高到低进行排名，筛选出排名靠前的预设个数的基金放入投资池内供适配。
  18. 根据权利要求17所述的存储介质，其中，所述在金融交易系统中设置配置文件时，使得一个或多个所述处理器执行以下步骤：
    在所述基金交易系统中预设配置界面，通过所述配置界面获取上传给所述基金交易系统的配置文件，并将所述配置文件存储在所述基金交易系统的数据库中。
  19. 根据权利要求17所述的存储介质，其中，所述调用所述得分计算脚本时，使得一个或多个所述处理器执行以下步骤：
    按一个季度为时间段，定时调用所述得分计算脚本后，筛选出所述排名靠前的预设个数的基金放入投资池内供适配。
  20. 根据权利要求17所述的存储介质，其中，所述将筛选出的所述基金在所述二级因子组内求等权重值时，使得一个或多个所述处理器执行如下计算公式：
    Figure PCTCN2018106590-appb-100004
    其中，N为所述二级因子组中基金的个数。
PCT/CN2018/106590 2018-06-25 2018-09-20 基金适配方法、系统、计算机设备和存储介质 Ceased WO2020000701A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810656996.X 2018-06-25
CN201810656996.XA CN108961063A (zh) 2018-06-25 2018-06-25 基金适配方法、系统、计算机设备和存储介质

Publications (1)

Publication Number Publication Date
WO2020000701A1 true WO2020000701A1 (zh) 2020-01-02

Family

ID=64486494

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/106590 Ceased WO2020000701A1 (zh) 2018-06-25 2018-09-20 基金适配方法、系统、计算机设备和存储介质

Country Status (2)

Country Link
CN (1) CN108961063A (zh)
WO (1) WO2020000701A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110059896A (zh) * 2019-05-15 2019-07-26 浙江科技学院 一种基于强化学习的股票预测方法及系统
CN115809763A (zh) * 2022-08-03 2023-03-17 联通(广东)产业互联网有限公司 一种监室排序智能分配的全周期管理方法、系统和计算机设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104376197A (zh) * 2014-10-29 2015-02-25 中国建设银行股份有限公司 一种参数风险评估的方法及装置
CN107563892A (zh) * 2017-05-31 2018-01-09 中科钢研节能科技有限公司 一种共建融资对象的分析方法与分析系统
CN107833137A (zh) * 2017-11-03 2018-03-23 上海宽全智能科技有限公司 基于多目标优化的量化交易策略生成方法及装置、设备和存储介质
CN108122166A (zh) * 2017-12-19 2018-06-05 苏州梧桐汇智软件科技有限责任公司 采用智能量化金融技术对股票进行归类和评分的方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104376197A (zh) * 2014-10-29 2015-02-25 中国建设银行股份有限公司 一种参数风险评估的方法及装置
CN107563892A (zh) * 2017-05-31 2018-01-09 中科钢研节能科技有限公司 一种共建融资对象的分析方法与分析系统
CN107833137A (zh) * 2017-11-03 2018-03-23 上海宽全智能科技有限公司 基于多目标优化的量化交易策略生成方法及装置、设备和存储介质
CN108122166A (zh) * 2017-12-19 2018-06-05 苏州梧桐汇智软件科技有限责任公司 采用智能量化金融技术对股票进行归类和评分的方法

Also Published As

Publication number Publication date
CN108961063A (zh) 2018-12-07

Similar Documents

Publication Publication Date Title
Korinek et al. From sudden stops to fisherian deflation: Quantitative theory and policy
AU2006252168B2 (en) System and method for processing composite trading orders
US7873565B2 (en) Composite trading order processing
CN112074860A (zh) 一种编译资产的投资组合的计算机实现的方法
US8407128B2 (en) Systems and methods for foreign exchange risk management
Ciganovic et al. Forecasting cryptocurrencies log-returns: a LASSO-VAR and sentiment approach
WO2020000701A1 (zh) 基金适配方法、系统、计算机设备和存储介质
Hamarat The efficiency of participation and conventional banking in Turkiye: a stochastic frontier approach
KR20160028701A (ko) 실시간 주식종목 추천시스템
JP6381844B1 (ja) 時間の経過につれて価値が変動するアセットを積み立てるためのコンピュータシステム、方法、および、プログラム
US20190385234A1 (en) System, Method, and Device for Autonomous Fund Management by Computer-Based Algorithms
TWI776370B (zh) 對於基金商品的投資風險評分方法及系統
KR20150124058A (ko) 종목 투자 가이드 시스템 및 그 방법
Polkowska Impact of costs on legal risk of CHF foreign currency mortgage loans in shaping financial performance of selected banks
LoGrasso Could ChatGPT Have Earned Abnormal Returns?
CN116167861A (zh) 自动化资产配置策略生成方法及装置
CN112732866A (zh) 投资者情绪指数构建方法、异质性主体市场模拟方法、设备及介质
CN115269994B (zh) 一种项目推荐方法和系统
TWM648707U (zh) 除權除息股票分析系統
JP2003067580A (ja) リスクヘッジ方法及びリスクヘッジシステム
Mohamed et al. Appraisal of Agricultural Policy in the Sudan
CN120355517A (zh) 一种面向风险对冲的投资组合构建方法、装置及存储介质
CN114913010A (zh) 量化投资方法及装置
Hilpert et al. The effect of secondary markets on equity-linked life insurance with surrender guarantees
JP2017120547A (ja) コミットメントライン追加引出率推計モデル

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18924143

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18924143

Country of ref document: EP

Kind code of ref document: A1