KR20220120242A - A method and apparatus for determining the proportion of an asset portfolio based on an optimal parameter calculation algorithm - Google Patents

Abstract

An operating method of an asset portfolio service providing apparatus according to an embodiment of the present invention includes the steps of: determining initial portfolio weight and stock information; based on the initial portfolio weight and stock information, determining a first weight parameter based on yield variable calculation corresponding to stock data of a first time interval from the present to a preset past time point; determining a second proportion parameter for an optimal section in which the rate of return variable is calculated highest according to a greedy search operation while, based on the initial portfolio weight and item information, performing rolling analysis processing for each second time interval corresponding to item data within a recent period; and determining an optimal weight parameter corresponding to the initial portfolio by performing a combination calculation of the first weight parameter and the second weight parameter. Optimization of asset portfolio proportion determination reflecting reality can be performed.

Description

A method and apparatus for determining the proportion of an asset portfolio based on an optimal parameter calculation algorithm}

The present invention relates to a method and apparatus for determining the weight of an asset portfolio. More specifically, the present invention relates to a method and apparatus for determining an asset portfolio weight based on an optimal parameter calculation algorithm.

Customers who have signed a contract for the management of financial assets with a financial company seek high returns and expect stability at the same time We offer a variety of products that you can pursue for profit.

As a method for efficiently allocating these products, portfolio selection theory is widely used. According to the portfolio selection theory, which is one of the strategies for effective asset investment, for asset investment, if the expected return is the same, a combination of products in the asset class with low risk is selected, and if the risk is the same, the expected return is higher. It is possible to combine products from asset class families.

In some cases, such portfolio preparation is done by individual investors, but in general, it is recommended to select a portfolio after receiving professional advice from a branch employee. However, in terms of time-consuming and difficult to secure reliability for such professional consultation, an automated portfolio creation tool such as a robo-advisor has been recently proposed.

Accordingly, current portfolio automation tools, such as robo-advisors, which have recently emerged, analyze customer propensities through questionnaires, evaluate and analyze them, and automatically suggest optimal portfolio models. In these current portfolio automation tools, a commonly used asset allocation algorithm employs a weighting method based on historical data analysis based on the portfolio theory of Harry Markowitz, who won the Nobel Prize in Economics in 1990.

Portfolio theory estimates a covariance matrix of data for a certain period of time in the past based on the expected return of a portfolio combining individual investment assets, and determines a diversified investment portfolio that can approach an efficient investment line based on its correlation coefficient. In particular, there is a problem in that the results vary greatly depending on how the expected return and the covariance matrix are estimated.

In particular, applying the model of portfolio theory involves estimating past data and assuming that the parameters will repeat or fit well in the future, and various problems that do not match reality occur.

For example, in the case of portfolio theory, there is a problem (not robust) in which the results of asset allocation are highly fluctuating in a highly volatile section, and the weight of asset allocation can be concentrated only in stocks with good short-term performance (bias). There is a problem, and since the expected return is estimated to be negative in the falling section, it may represent a result that is different from reality, such as deriving a logically meaningless result.

Accordingly, the need for an asset portfolio weighting method to provide a more realistic and accurate diversification solution is emerging.

The present invention has been devised to solve the above problems, and it is not estimating only with simple past data, but calculating the optimal parameters of a realistic and accurate distribution model based on a prior search, and determining the combination of the weight of the asset portfolio based on this An object of the present invention is to provide a method and device for determining the weight of the asset portfolio that can optimize the determination of the weight of the asset portfolio that reflects the reality.

A method according to an embodiment of the present invention for solving the above-described problems, the method of operating an asset portfolio service providing apparatus, the method comprising: determining an initial portfolio weight and stock information; Based on the initial portfolio weight and stock information, based on the return variable calculation corresponding to stock data of a first time interval from the present to a preset past time, determining a first weight parameter; Based on the initial portfolio weight and stock information, while performing rolling analysis processing for each second time section corresponding to stock data within a recent certain period range, the return variable according to a greedy search operation Determining a second specific gravity parameter for the optimal section in which is calculated the highest; and combining the first weighting parameter and the second weighting parameter to determine an optimal weighting parameter corresponding to the initial portfolio.

In addition, an apparatus according to an embodiment of the present invention for solving the above problems, in the asset portfolio service providing apparatus, the asset allocation engine management unit for determining the initial portfolio weight and stock information; Based on the initial portfolio weight and stock information, based on a return variable calculation corresponding to stock data of a first time interval from the present to a preset past time, a first weight parameter determining unit for determining a first weight parameter; Based on the initial portfolio weight and stock information, while performing rolling analysis processing for each second time section corresponding to stock data within a recent certain period range, the return variable according to a greedy search operation a second specific gravity parameter determining unit for determining a second specific gravity parameter for an optimal section in which α is the highest; and an optimal parameter combination calculator configured to calculate an optimal weighting parameter corresponding to the initial portfolio by combining the first weighting parameter and the second weighting parameter.

On the other hand, in the method according to an embodiment of the present invention for solving the above-described problems, a computer program stored in a computer-readable medium for executing the method in a computer, and a recording medium in which the program is stored. can

According to an embodiment of the present invention, an asset capable of optimizing the determination of the proportion of the asset portfolio reflecting the reality by calculating the optimal parameter of a realistic and accurate distribution model based on the prior search and determining the proportion of the asset portfolio based on the combination according to the weight. A method and apparatus for determining portfolio weighting may be provided.

1 is a conceptual diagram schematically illustrating an entire system according to an embodiment of the present invention.
2 is a block diagram specifically illustrating a configuration of a service providing apparatus according to an embodiment of the present invention.
3 is a block diagram illustrating a user terminal device according to an embodiment of the present invention.
4 is a block diagram illustrating in more detail a portfolio weight determining unit according to an embodiment of the present invention.
5 is a flowchart illustrating a method of operating a service providing apparatus according to an embodiment of the present invention.
6 is a view for explaining a test result according to an embodiment of the present invention.

The following is merely illustrative of the principles of the invention. Therefore, those skilled in the art will be able to devise various devices that, although not explicitly described or shown herein, embody the principles of the present invention and are included within the spirit and scope of the present invention. Further, it is to be understood that all conditional terms and examples listed herein are, in principle, expressly intended solely for the purpose of enabling the concept of the present invention to be understood, and not limited to such specifically enumerated embodiments and states. should be

Moreover, it is to be understood that all detailed description reciting the principles, aspects, and embodiments of the invention, as well as specific embodiments, are intended to cover structural and functional equivalents of such matters. It is also to be understood that such equivalents include not only currently known equivalents, but also equivalents developed in the future, i.e., all devices invented to perform the same function, regardless of structure.

Thus, for example, the block diagrams herein are to be understood as representing conceptual views of illustrative circuitry embodying the principles of the invention. Similarly, all flowcharts, state transition diagrams, pseudo code, etc. may be tangibly embodied on a computer-readable medium and be understood to represent various processes performed by a computer or processor, whether or not a computer or processor is explicitly shown. should be

In addition, the clear use of terms presented as processor, control or similar concepts should not be construed as exclusively referring to hardware having the ability to execute software, and without limitation, digital signal processor (DSP) hardware, ROM for storing software. It should be understood to implicitly include (ROM), RAM (RAM) and non-volatile memory. Other common hardware may also be included.

The above-described objects, features and advantages will become more apparent through the following detailed description in relation to the accompanying drawings, and accordingly, those of ordinary skill in the art to which the present invention pertains can easily implement the technical idea of the present invention. There will be. In addition, in the description of the present invention, when it is determined that a detailed description of a known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted.

Hereinafter, a preferred embodiment according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a conceptual diagram schematically illustrating an entire system according to an embodiment of the present invention.

Referring to FIG. 1 , the entire system according to an embodiment of the present invention includes a service providing apparatus 100 , a user terminal 200 , and a financial server 300 .

The entire system may be interconnected through a communication network including a wired/wireless Internet network, respectively. In addition, a network for securely connecting each communication may be configured between the service providing apparatus 100 , the user terminal 200 , and the financial server 300 . In addition, each of the service providing apparatus 100 , the user terminal 200 , and the financial server 300 may each include a communication module for transmitting and receiving data by accessing the communication network.

The communication network includes, for example, an open Internet, a wired Internet network including a closed intranet, a wireless Internet communication network interlocked with a mobile communication network, a protocol such as Transmission Control Protocol (TCP)/Internet Protocol (IP), and User Datagram Protocol (UDP). A computer network capable of various data communication including data communication based on the data may be exemplified.

In addition, the service providing apparatus 100 may process one or more processes for providing an asset allocation service according to an embodiment of the present invention to the financial server 300 according to a request of the user terminal 200 , An asset allocation service may be provided to the user terminal 200 .

Here, the service providing apparatus 100 may provide an asset allocation solution analysis and asset allocation processing service according to the request of the user terminal 200 . In particular, the service providing apparatus 100 according to an embodiment of the present invention enables the creation of a user-customized model, and in response to asset allocation according to the currently applied portfolio, it can process model simulation and real-time reapplication according to the change setting. have.

In addition, the service providing device 100 provides the generated customized asset allocation model to be uploaded to the website, enabling relevant information to be provided online as well as offline, and all customized asset allocation models are algorithmized and automatically may include automated financial information processing processes that enable transaction processing.

To this end, the service providing apparatus 100 according to an embodiment of the present invention acquires user analysis information based on user input information input to a user terminal, and determines asset setting information corresponding to the user analysis information, The asset setting information is applied to an asset allocation engine based on pre-stored model information to generate initial portfolio information, provides the initial portfolio information to the user terminal, and user customization corresponding to the initial portfolio information from the user terminal By obtaining information and providing real-time simulation information using the user-customized asset allocation model determined based on the user-customized information to the user terminal, a user-customized asset allocation model can be created, and the real-time simulation information for this can be provided to the user terminal ( 200) can be provided.

Here, the asset allocation engine may include a process capable of optimizing the portfolio weight based on the initial portfolio information. To this end, when the initial portfolio weight and stock information are determined in the asset allocation engine, the return variable is calculated based on the initial portfolio weight and stock information, corresponding to the stock data of the first time period from the present to the preset past time. Based on the determination of the first weight parameter, and on the basis of the initial portfolio weight and item information, while performing a second time interval rolling analysis processing corresponding to item data within a recent certain period range, According to a greedy search operation, a second weighting parameter is determined for the optimal section in which the yield variable is calculated the highest, and the first weighting parameter and the second weighting parameter are combined and calculated to correspond to the initial portfolio It is possible to determine the optimal specific gravity parameter.

According to the portfolio weight optimization calculation of the asset allocation engine, the asset allocation engine according to the embodiment of the present invention can determine a realistic and appropriate asset allocation weight by comprehensively considering the recent market conditions and the existing market conditions. It makes it possible to build an asset allocation model that is optimized for simulation.

In addition, the service providing apparatus 100 may include an asset allocation engine that processes storage and management of one or more asset allocation models, and a database for storing user information, asset setting information, trading registration information, and simulation information, and more A detailed configuration will be described later with reference to FIG. 2 .

In addition, the service providing apparatus 100 may communicate with the financial server 300 to perform financial processing according to the automatic trading service on the asset setting information. The financial server 300 may correspond to one or more financial institution servers, for example, a bank server or securities company server that processes asset transfer or transaction information to an investment target account according to the request of the service providing device 100 . can be exemplified.

In particular, the service providing apparatus 100 according to an embodiment of the present invention processes the simulation of the user-customized asset allocation model and transaction information with the financial server 300 according to the application of the user-customized asset allocation model, thereby automatically based on the simulation. We can provide trading services.

The simulation and automatic trading service of the service providing apparatus 100 may be executed by a robo-advisor for each item processed in response to each product item included in the asset distribution model as the user-customized asset distribution model is determined.

To this end, the service providing apparatus 100 may compose the robo-advisors for each item in a complex manner according to the determination of the asset allocation model, and the complex construction process may be automated. Accordingly, the service providing apparatus 100 according to an embodiment of the present invention can function as a robo-advisor of a higher concept using the robo-advisor, and facilitates portfolio simulation and re-application processing.

On the other hand, the user terminal 200 receives user information, transmits and receives data to and from the asset distribution service providing apparatus, and provides an asset management service based on the user information input and communication with the asset distribution service providing apparatus. It may be a terminal. For example, the user terminal 200 is an electronic device in which a service function according to an embodiment of the present invention can be installed through an application, for example, a smartphone, a laptop computer, a tablet PC, a kiosk, and the like.

In addition, the user according to an embodiment of the present invention may be a user provided with an asset management service. For example, the user may be a direct investor as a user of the user terminal 200 , but various users who receive asset management services such as a designer who designs an investment portfolio and an investment consultant through the user terminal 200 may be exemplified. have.

In addition, the user terminal 200 may process the information provided from the asset allocation service providing apparatus and output it through a display interface, and may include one or more output units (not shown) for this purpose. The output unit may include an interface for providing information to the user. For example, the output may have one or more display interfaces.

2 is a block diagram for explaining in more detail an apparatus for providing an asset allocation service according to an embodiment of the present invention.

Referring to FIG. 2 , the service providing apparatus 100 according to an embodiment of the present invention includes a communication unit 105 , an asset allocation engine management unit 120 , an item setting information management unit 130 , a user information analysis unit 140 , and a simulation. It includes a processing unit 150 , a web site management unit 160 , an automatic trading processing unit 170 , a portfolio reapplication processing unit 190 , and a database management unit 180 .

The communication unit 105 includes one or more communication modules for communicating with the financial server 300 or the user terminal 200 , or communicating with a network in which the financial server 300 or the user terminal 200 is located.

Here, the communication unit 105 is a type connected by a wireless or wired method through a local area network (LAN) and an Internet network, a type connected through a USB (Universal Serial Bus) port, and a mobile communication network such as 3G or 4G. It is possible to connect through a short-range wireless communication method such as NFC (Near Field Communication, RFID (Radio Frequency Identification), Wi-Fi, etc.).

In addition, the communication unit 105 may receive user registration information for providing an asset management and asset distribution service from the user terminal 200 , and the user information analysis unit 140 receives user information according to the received user registration information. A registration process may be performed, and a user management process corresponding to the registered user information may be performed.

In particular, the user information analysis unit 140 may provide a user analysis interface for user analysis to the user terminal 200, and may obtain user analysis information by analyzing user input information corresponding to the user analysis interface. have.

The item setting information management unit 130 may receive asset information from the user terminal 200 , and may obtain, store, and manage item setting information corresponding to the user asset information. Here, the stock setting information may include information on financial assets that the user wants to invest in and asset group allocation information according to the user's selection, which may be obtained from user analysis information.

In addition, the asset allocation engine management unit 120 may determine the initial portfolio by applying the user analysis information and the item setting information to the asset allocation engine according to the preset asset allocation model information.

For example, the asset allocation engine management unit 120 may drive the asset allocation engine by applying the user analysis information and the item setting information to the portfolio determination algorithm according to the preset asset allocation model, and the asset including the initial portfolio. Distribution model information can be output. To this end, the asset allocation engine management unit 120 may store and manage one or more asset allocation model engine information driven based on user analysis information.

In addition, asset allocation model information including the initial portfolio may be provided to the user terminal 200 . Here, the initial portfolio information may include a portfolio determined by an asset allocation model determined in response to the user analysis information according to the operation of the asset allocation engine management unit 120 among one or more previously known standard asset allocation modules.

For example, the asset allocation model of the initial portfolio information may be determined according to the asset investment propensity according to user analysis, and may include an asset allocation ratio for each asset type according to the asset investment propensity. For example, asset types may include ETFs by stock, deposits/liquidity funds, domestic bonds, domestic stocks, overseas stocks, foreign bonds, alternative investment funds, structured bonds, and bancassurance insurance. The asset allocation ratio can be determined. Accordingly, the initial portfolio information may include investment ratio information corresponding to stocks for each asset type, and asset allocation model information corresponding thereto may be visualized in the form of a house view and provided to the user terminal 200 .

Meanwhile, in determining the asset allocation model, the portfolio weight determination unit 125 may perform an optimization process for determining the weight between stocks in the asset allocation model. It calculates the optimal parameters of a realistic and accurate distribution model based on a prior search and determines the weight of the asset portfolio based on the combination according to the weight, thereby optimizing the determination of the weight of the asset portfolio reflecting the reality.

To this end, when the initial portfolio weight and item information are determined, the portfolio weight determination unit 125 corresponds to item data of the first time period from the present to a preset past time point based on the initial portfolio weight and item information. Based on the return variable calculation to determine a first weight parameter, and based on the initial portfolio weight and item information, the second time interval rolling analysis processing corresponding to item data within a recent certain period range while performing a greedy (Greedy) search operation to determine a second weighting parameter for the optimal section in which the yield variable is calculated the highest, and combining the first weighting parameter and the second weighting parameter, An optimal weighting parameter corresponding to the initial portfolio may be determined.

These optimal weight parameters are calculated to assume an appropriate expected return, and are not simply predicted from the results of a specific period, but are optimized to assume a reasonable expected return. It can be determined by an optimization process that determines the appropriate observation period and output variables to find and calculate from.

Since this is not simply estimated from data of the past period such as the last one year or six months as in the conventional method, it reduces the inaccuracy of parameters due to high volatility sections, and also reduces the bias effect that is concentrated only on stocks with high performance in the short term. In particular, it is possible to build an optimized asset distribution model that can prevent errors such as logically meaningless results due to negative expected returns in a falling section.

Accordingly, the user terminal 200 may output the initial portfolio information and an asset distribution model based thereon through the portfolio interface. In addition, the user terminal 200 may receive user customized information from the user, and the inputted customized information may be transmitted to the service providing apparatus 100 . For example, the initial portfolio information may be an optimal investment ratio determined using the above-described optimization parameter, but the user may request a personalized and customized stock setting. Accordingly, according to an embodiment of the present invention, the user terminal 200 may receive user-customized information for adding or changing a customized asset for each item, and this user-customized information may be transmitted to the service providing apparatus 100 . .

In addition, the asset allocation engine management unit 120 may change the initial portfolio model to a user-customized asset allocation model based on the user-customized information, and the changed user-customized asset allocation model information may be transmitted to the simulation processing unit 150 . have.

The simulation processing unit 150 may first process a model simulation based on initial portfolio information and provide the processed result information to the user terminal 200 . Here, the financial information simulated according to the initial portfolio information may include simulation results based on past financial information, and an average annual expected return, annual volatility, maximum drop, Sharpe ratio, and the like may be displayed.

In addition, when the user-customized asset allocation model is determined, the simulation processing unit 150 may provide real-time simulation processing according to the user-customized asset allocation model, and the simulation result may be provided to the user terminal 200 . Here, the user terminal 200 may compare and provide the simulation result of the initial portfolio model and the simulation result of the user-customized asset allocation model, thereby facilitating the user's performance evaluation and the like.

On the other hand, the user terminal 200 may request a user trading registration corresponding to the user-customized asset allocation model, and the automatic trading processing unit 170, according to the user trading registration request, includes user account information and initial asset information. Receive user trading registration information, and execute an automatic trading process based on the user-customized asset allocation model.

Accordingly, the trading processing unit 170 may provide the current asset allocation portfolio monitoring information according to the automatic trading process to the user terminal 200 . Such monitoring information may be provided in such a way that the current asset allocation portfolio model information is provided through the portfolio interface of the user terminal 200 .

Meanwhile, the portfolio reapplication processing unit 190 may receive change setting information corresponding to the current asset allocation portfolio model information from the user terminal 200 .

To this end, the user terminal 200 may receive user change setting information through the portfolio interface, the engine management unit 120 may generate a change portfolio model based on the change setting information, and the simulation processing unit 150 may By processing the simulation based on the changed portfolio model, real-time simulation result comparison information with the current asset allocation portfolio model may be provided to the user terminal 200 . Accordingly, the user terminal 200 may check the simulation result comparison and input a real-time portfolio model change request.

The portfolio reapplication processing unit 190 may perform reapplication processing of the changed portfolio model corresponding to the current account and asset of the user trading registration information, based on the real-time portfolio model change request.

Here, the portfolio reapplication processing unit 190 acquires current account information and current asset information corresponding to the previously registered user trading registration information without adding or changing user trading registration information, and obtaining the current account information and the current Automatic change processing of the user trading registration information for re-application processing of asset information from the current portfolio model to the changed portfolio model may be performed.

Accordingly, the automatic trading processing unit 170 may perform the automatic trading process based on the automatically changed user trading registration information and the changed portfolio model.

Meanwhile, the web site management unit 160 may create and upload one or more user-customized asset allocation models for each user, and may provide and operate a web site that can be shared with each other. To this end, the website management unit 160 may register the user-customized asset allocation model by matching it with user information, and process the web page management so that the registered user-customized asset allocation model can be shared with another second user terminal. can

For example, the website management unit 160 may share one or more recommended user-customized asset allocation model information to the second user terminal according to the user analysis information of the second user terminal, and thus, It is easy to set up and change the asset allocation model using the customized asset allocation model information.

Meanwhile, the controller 110 controls the overall operation of each of the above-described components, and may include one or more microprocessors for this purpose.

In addition, the database management unit 180 includes model information for applying the aforementioned asset allocation engine, item setting information for each user, user analysis information, simulation processing information, website management information, automatic trading processing information, portfolio re-application processing information, etc. It handles the storage and management of , and may include one or more storage media for this.

3 is a block diagram illustrating a user terminal according to an embodiment of the present invention in more detail.

Referring to FIG. 3 , the user terminal 200 according to an embodiment of the present invention includes an asset management service unit 210 , a communication unit 220 , a control unit 230 , a user input unit 240 , and a memory 250 . do.

The communication unit 220 may include one or more modules that enable communication between the user terminal 200 and the service providing apparatus 100 or a network in which the service providing apparatus 100 is located. For example, the communication unit 220 may include a mobile communication module, a wireless Internet module, a short-range communication module, or a wired communication module.

The user input unit 240 may include one or more interface modules for receiving a user input. More specifically, the user input unit 240 is the service providing device 100 for user information, user input information for user analysis, user customized information for creating a user-customized asset allocation model, model change request information, automatic trading registration information, etc. A variety of information can be input and processed.

The control unit 230 may control the general operation of the general terminal 200 , and in particular may control various asset distribution service operations through the asset management service unit 210 .

The asset management service unit 210 may provide various asset management services through communication with the service providing apparatus 100 under the control of the controller 230 .

To this end, the asset management service unit 210 includes a propensity analysis module for outputting the user analysis information according to user input information, a house view module for outputting the asset setting information in a house view method, and the user-customized asset allocation model information an asset allocation module for outputting , a simulation module for outputting the real-time simulation information; and when an automatic trading process based on the user-customized asset allocation model is executed by receiving user trading registration information including the user account information and initial asset information from the asset allocation service providing device, the current asset allocation according to the automatic trading process It may include an automatic trading module that outputs portfolio monitoring information.

Accordingly, the user terminal 200 may be provided with the asset management service processed by the asset management service unit 210 and the asset distribution service provided from the service providing device 100 , and various service information processed is provided by the control unit. It may be output as an image through the output unit by 230 .

In addition, the memory 250 may include one or more storage media for storing input/output information for a process for controlling the controller 230 and an operation of each component.

4 is a block diagram illustrating in more detail a portfolio weight determining unit according to an embodiment of the present invention.

Referring to FIG. 4 , the portfolio weight determining unit 125 according to an embodiment of the present invention includes a first weighting parameter determining unit 1251 , a second weighting parameter determining unit 1252 , and an optimal parameter combination calculating unit 1253 . include

First, as described above, when the portfolio weight determination unit 125 determines the initial portfolio weight and item information, based on the initial portfolio weight and item information, items in the first time interval from the present to a preset past time point Based on the calculation of the return variable corresponding to the data, the first weight parameter is determined, and on the basis of the initial portfolio weight and item information, the second time section corresponding to the item data within a recent certain period range is rolled (Rolling). ) while performing analysis processing, according to a greedy search operation, a second weight parameter is determined for an optimal section in which the yield variable is calculated the highest, and the first weight parameter and the second weight parameter are combined and calculated Thus, to determine the optimal weight parameter corresponding to the initial portfolio, for this purpose, first, the first weight parameter determining unit 1251, for maximizing the Sharpe Ratio variable for the first fixed time period A first optimal portfolio distribution parameter is determined.

Here, the Sharpe ratio represents the risk-adjusted return of the portfolio, and according to the Sharpe ratio setting, an excess return over risk-free assets for taking additional risks may be determined, and an asset distribution model may be determined accordingly.

Then, the second specific gravity parameter determining unit 1252 maximizes the Sharpe Ratio variable for the second time period that is varied within the predetermined period range according to the rolling analysis processing, and is within the predetermined period range. A second optimal portfolio distribution parameter in the optimal interval in which the Sharpe ratio variable in is maximum is determined.

Then, the optimal parameter combination operation unit 1253 multiplies the first optimal portfolio distribution parameter by a first weight, multiplies the second optimal portfolio distribution parameter by a second weight, and the multiplied first By summing the optimal portfolio distribution parameter and the multiplied second optimal portfolio distribution parameter, an optimal weight parameter corresponding to the initial portfolio may be calculated.

Here, the sum of the first weight and the second weight may be 1, preferably, the first weight may be 0.8, and the second weight may be 0.2. The basis for calculating the weight will be described later based on the test result shown in FIG. 6 .

More specifically, for example, the first optimal portfolio distribution parameter is the Sharpe ratio ( Sharpe ratio) may include a variable, and the first specific gravity parameter determiner 1251 may calculate it and set it to A.

Then, the second weight parameter determining unit 1252, in response to stock data within a recent predetermined period range (past N years to N + 6 months, etc.), to determine the second optimal portfolio distribution parameter, the N A rolling analysis process for each second time section that sequentially or selectively varies up to a specific time point may be performed.

Then, the second weight parameter determining unit 1252 performs a greedy search operation based on the return variable calculated according to the rolling analysis process, and the second weight parameter corresponding to the optimal section in which the return variable is the highest. 2 The specific gravity parameter can be determined as B.

Here, the greedy search operation is a result of applying the preset Sharpe ratio variable for each time section variable according to the rolling analysis, and the calculation day for predicting and determining the Sharpe ratio variable of the section where the yield is optimized as the second weight parameter can

Here, the rolling-based greedy search operation for calculating the section to be optimized is a method of determining the final optimal value by iteratively calculating a value expected to be optimal at every selection time, and is an embodiment of the present invention The second weighting parameter determining unit 1252 according to It may be processed as an operation of determining a Saffron ratio variable corresponding to the section as an optimal parameter of the entire section data and outputting it as the second specific gravity parameter.

Accordingly, the second weight parameter determining unit 1252 performs the greedy search operation for deriving the value of the Sharpe ratio parameter that maximizes the yield variable based on the stock data as an optimal parameter, the plurality of rolling sections Calculating the maximum yield variable compared to each Sharp ratio parameter according to the application of a plurality of sharp ratio parameters for each, and determining the rolling section maximizing the yield variable while sequentially searching the plurality of rolling sections as the optimal section, search of the rolling section The optimal section is updated according to whether or not the yield variable is maximized at each time point, and an optimal Saffron ratio parameter corresponding to the last updated optimal section is determined as the optimal parameter of the entire section data of the stock data according to the end of rolling, and the determined optimal The parameter can be output as the second specific gravity parameter.

Since such a greedy search operation may not index all time intervals in detail, it does not always yield the best results, but it can quickly calculate the second specific gravity parameter, and enables efficient processing for rapid modeling processing. . Such a greedy algorithm is an algorithm used when determining a set of optimal solutions of a part as a solution to the whole problem, and can be a method of quickly obtaining an approximate optimal value by abbreviating numerical operations in too many cases. The second weight parameter calculation unit 1252 according to the example uses the greedy algorithm for the search process of the rolling section from which the maximum yield is derived, thereby rapidly calculating the second weight parameter.

Then, the optimal parameter combination operation unit 1253 multiplies the first weight W1 set in A, multiplies the second weight W2 set in B, and sums them to obtain the final combined weight value of the sharp ratio (Sharpe ratio) can be determined based on a variable. Accordingly, the finally determined optimal specific gravity parameter C may be determined by a combination operation such as C = W1 * A + W2 * B.

Here, W1 and W2 may be random variables determined by a user, and their sum may be 1. Accordingly, the user can determine and apply an appropriate weight more suitable for reality by setting and testing each weight. Among the results calculated according to an embodiment of the present invention, the most preferable values are a case in which W1 is 0.8 and W2 is 0.2, which will be described later with reference to FIG. 6 .

5 is a flowchart illustrating a method of operating a service providing apparatus according to an embodiment of the present invention.

Referring to FIG. 5 , when the service providing apparatus 100 according to an embodiment of the present invention determines an initial portfolio weight and item ( S101 ), based on the initial portfolio weight and item, from the present to a preset past time point Based on the Sharpe ratio variable corresponding to the item data of the first time interval, a first weight parameter is determined (S103).

In addition, the service providing apparatus 100 performs the rolling analysis process for each second time section corresponding to the item data for the last N years, and the second for the optimal section in which the sharp ratio variable is calculated the highest according to the greedy search operation. Determine the specific gravity parameter (S105).

Thereafter, the service providing apparatus 100 determines the optimal weight parameter by applying a combination of the first weight corresponding to the first weight parameter and the second weight corresponding to the second weight parameter ( S107 ).

Then, the service providing apparatus 100 procures an asset distribution model based on the determined optimal weight parameter, and performs simulation, automatic trading processing, or redistribution service processing based thereon ( S109 ).

6 is a view for explaining a test result according to an embodiment of the present invention.

Referring to FIG. 6 , the test according to the embodiment of the present invention predicts performance ( Performance) and the test results of comparative analysis are shown.

First, the second optimal weight parameter-based model table is based on the GICS index-based stock data of the previous year, rolling the 250-day time interval, and the parameter of the interval in which the maximum Sharpe ratio parameter is calculated. As calculated by the second optimal weight parameter, each item corresponding to WMI500 Energy, WMI500 Materials, WMI500 Industrial Goods, WMI500 Consumer Consumer Goods, WMI500 Consumer Essentials, WMI500 Health Care, WMI500 Finance, WMI500 Information Technology, WMI500 Communication Service and WMI500 Utility And it is the result of statistical analysis of performance by calculating the optimized value of the SAF ratio for each section and applying it from the first half of 2016 to the first half of 2020, respectively.

Here, opt means the result optimized by the second optimal weight parameter, and the comparison between the case of equal weight (equal) and the case of the actual stock data (kospi) compared to this are calculated numerically As a result, ret represents the return, and sr represents the Sharpe Ratio.

And, in the first optimal weight parameter-based model table, a parameter representing a high Sharp index was calculated from stock data of the previous five years of the same GICS index-based data, and the average applied value was used by selecting the TOP 10 section. Similarly, the maximum value of the SAF ratio based on each stock corresponding to WMI500 Energy, WMI500 Materials, WMI500 Industrial Goods, WMI500 Consumer Consumer Goods, WMI500 Consumer Staples, WMI500 Health Care, WMI500 Finance, WMI500 Information Technology, WMI500 Communication Services and WMI500 Utilities. It is the result of statistical analysis of performance by calculating and applying it from the first half of 2016 to the first half of 2020.

Here, opt means the result optimized by the first optimal weight parameter, and the comparison between the case of equal weight (equal) and the case of the actual stock data (kospi) compared to this are numerically calculated. As a result, ret represents the return, and sr represents the Sharpe Ratio.

In addition, the weight combination model performance comparison table compares performances calculated by applying the weighting calculation to the first optimal weight parameter-based model and the second optimal weight parameter-based model, and the performance according to the ratio of each weight (0.0 to 1.0) shows the improvement effect of

As shown in FIG. 6 , when comparing the average performance (mean) of the difference values corresponding to the return (ret) and the Sharpe ratio (sr) of each model numerically, the effect is different according to each weight, but the first It was confirmed that, compared to the optimal specific gravity parameter model and the second optimal specific gravity parameter model, the variable volatility (Robust) is reduced, and it was confirmed that an optimized model showing a high Sharpe ratio (sr) can be configured.

In particular, it is confirmed that the section where the SAF ratio is maximized compared to the actual kospi stock weight data or equal weight data appears in the 0.8mix case, in which the second weight is set to 0.8 and the first weight is set to 0.2. Therefore, it can be confirmed that an optimized combination operation parameter can be calculated according to the weight setting. Therefore, according to the current test result, in general stock data, it may be preferable that the first weight and the second weight are 0.2: 0.8, and the user adjusts these weights and decides to subdivide the weights more suitable for the current situation or data. You can also adjust it.

The method according to the present invention described above may be produced as a program to be executed by a computer and stored in a computer-readable recording medium. Examples of the computer-readable recording medium include ROM, RAM, CD-ROM, magnetic tape. , floppy disks, and optical data storage devices.

The computer-readable recording medium is distributed in a network-connected computer system, so that the computer-readable code can be stored and executed in a distributed manner. In addition, functional programs, codes, and code segments for implementing the method can be easily inferred by programmers in the art to which the present invention pertains.

In addition, although preferred embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above, and the technical field to which the present invention belongs without departing from the gist of the present invention as claimed in the claims Various modifications may be made by those of ordinary skill in the art, and these modifications should not be individually understood from the technical spirit or prospect of the present invention.

Claims (18)

A method of operating an asset portfolio service providing device, comprising:
determining the initial portfolio weight and stock information;
Based on the initial portfolio weight and stock information, based on the return variable calculation corresponding to stock data of a first time interval from the present to a preset past time, determining a first weight parameter;
Based on the initial portfolio weight and stock information, while performing rolling analysis processing for each second time section corresponding to stock data within a recent certain period range, the return variable according to a greedy search operation Determining a second specific gravity parameter for the optimal section in which is calculated the highest; and
Combining the first weighting parameter and the second weighting parameter to determine an optimal weighting parameter corresponding to the initial portfolio
How an asset portfolio service delivery device works. According to claim 1,
The first weight parameter includes a first optimal portfolio distribution parameter that maximizes a Sharpe Ratio variable for the first fixed time period
How an asset portfolio service delivery device works. 3. The method of claim 2,
The second specific gravity parameter is,
Maximizes the Sharpe Ratio variable for the second time period that varies within the predetermined period range according to the rolling analysis processing, and in the optimal period in which the Sharpe ratio variable within the predetermined period range is maximum including a second optimal portfolio distribution parameter of
How an asset portfolio service delivery device works. 4. The method of claim 3,
The determining step is
multiplying the first optimal portfolio distribution parameter by a first weight, multiplying the second optimal portfolio distribution parameter by a second weight, and multiplying the multiplied first optimal portfolio distribution parameter and the multiplied second By summing the optimal portfolio distribution parameters, calculating the optimal weight parameter corresponding to the initial portfolio
How an asset portfolio service delivery device works. 5. The method of claim 4,
The sum of the first weight and the second weight is 1
How an asset portfolio service delivery device works. 6. The method of claim 5,
The first weight is 0.8 and the second weight is 0.2
How an asset portfolio service delivery device works. According to claim 1,
Further comprising the step of performing the asset distribution model building process of the initial portfolio based on the determined optimal weight parameter
How an asset portfolio service delivery device works. 8. The method of claim 7,
Further comprising the step of performing a simulation process based on the asset distribution model
How an asset portfolio service delivery device works. In the asset portfolio service providing device,
Asset allocation engine management unit that determines initial portfolio weight and stock information;
Based on the initial portfolio weight and stock information, based on a return variable calculation corresponding to stock data of a first time interval from the present to a preset past time, a first weight parameter determining unit for determining a first weight parameter;
Based on the initial portfolio weight and stock information, while performing rolling analysis processing for each second time section corresponding to stock data within a recent certain period range, the return variable according to a greedy search operation a second specific gravity parameter determining unit for determining a second specific gravity parameter for an optimal section in which α is the highest; and
and an optimal parameter combination calculating unit configured to combine the first weighting parameter and the second weighting parameter to determine an optimum weighting parameter corresponding to the initial portfolio.
Asset Portfolio Service Delivery Device. 10. The method of claim 9,
The first weight parameter includes a first optimal portfolio distribution parameter that maximizes a Sharpe Ratio variable for the first fixed time period
Asset Portfolio Service Delivery Device. 11. The method of claim 10,
The second specific gravity parameter is,
Maximizes the Sharpe Ratio variable for the second time period that varies within the predetermined period range according to the rolling analysis processing, and in the optimal period in which the Sharpe ratio variable within the predetermined period range is maximum including a second optimal portfolio distribution parameter of
Asset Portfolio Service Delivery Device. 12. The method of claim 11,
The optimal parameter combination calculating unit,
multiplying a first weight corresponding to the first optimal portfolio distribution parameter, multiplying a second weight corresponding to the second optimal portfolio distribution parameter, and multiplying the multiplied first optimal portfolio distribution parameter and the first optimal portfolio distribution parameter 2 By summing the optimal portfolio distribution parameters, the optimal weight parameter corresponding to the initial portfolio is calculated.
Asset Portfolio Service Delivery Device. 13. The method of claim 12,
The sum of the first weight and the second weight is 1
Asset Portfolio Service Delivery Device. 14. The method of claim 13,
The first weight is 0.8 and the second weight is 0.2
Asset Portfolio Service Delivery Device. 10. The method of claim 9,
The asset allocation engine management unit,
Performing asset distribution model building processing of the initial portfolio based on the determined optimal weighting parameter based on the determined optimum weighting parameter
Asset Portfolio Service Delivery Device. 16. The method of claim 15,
The asset allocation engine management unit,
Performing simulation processing based on the redistributed asset portfolio
Asset Portfolio Service Delivery Device. 10. The method of claim 9,
The second specific gravity parameter determining unit,
In performing the greedy search operation for deriving the value of the Sharpe ratio parameter that maximizes the yield variable based on the stock data as the second weight parameter,
Calculate the maximum yield variable compared to each sharp ratio parameter according to the application of a plurality of sharp ratio parameters for each of the plurality of rolling sections, and determine the rolling section maximizing the yield variable while sequentially searching the plurality of rolling sections as the optimal section, The optimal section is updated according to whether or not the yield variable is maximized at each search time of the rolling section, and the optimal Saffron ratio parameter corresponding to the finally updated optimal section according to the end of rolling is determined as the optimal parameter of the data of the entire section of the item data. and outputting the determined optimal parameter as the second specific gravity parameter
Asset Portfolio Service Delivery Device. A computer program stored in a computer-readable medium for executing the method according to any one of claims 1 to 8 in a computer.

Images