Investment environment assessment system, investment environment assessment method, and investment environment assessment program

Abstract

This invention provides an investment environment assessment system, an investment environment assessment method, and an investment environment assessment program for efficiently preparing operational reports. [Solution] The support server 20 includes a user terminal and a control unit connected to the generating AI. The control unit 21 obtains information on the fund and the analysis period from the user terminal 10, obtains benchmarks related to the fund's target market, and identifies inflection points in the analysis period according to a resolution appropriate for the analysis period. The control unit 21 classifies the market phases between each identified inflection point, obtains comments on the fund's investment environment for each market phase from the AI ​​system 30, and outputs the comments and the basis for the comments to the user terminal 10.

Description

Technical Field

[0001] The present disclosure relates to an investment environment evaluation system, an investment environment evaluation method, and an investment environment evaluation program for evaluating an investment environment.

Background Art

[0002] A trust company for investment trusts creates an operation report to be delivered to investors every settlement / monthly. Therefore, a report creation system for assisting in creating the operation report has been considered (for example, Patent Document 1). The report creation system described in this document has one or more components including a module that outputs predetermined display contents based on data held in a data source in a plurality of report templates. A module corresponding to a component designated in the report template is called. Then, an operation report is generated by replacing the component designated in the report template with the output display content.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The fund settlement component of this patent document is a component that displays content uniquely determined in units of each settlement period for each fund. For example, this component includes the name of the settlement period, the settlement date, data on operation results for each settlement period, tables, graphs, comments from the fund manager, etc. However, it takes time to create comments on the investment environment and the like. Also, it is necessary to perform fact-checking on the operation comments.

Means for Solving the Problems

[0005] The investment environment evaluation system that solves the above problems comprises a user terminal and a control unit connected to a generating AI. The control unit acquires information on a fund and the analysis period from the user terminal, acquires benchmarks related to the fund's target market, identifies inflection points in the analysis period according to a resolution appropriate for the analysis period for the benchmarks, classifies the market phases between each identified inflection point, acquires comments on the fund's investment environment for each market phase from the generating AI, and outputs the comments and the basis for the comments to the user. [Effects of the Invention]

[0006] According to the present invention, reports can be prepared efficiently. [Brief explanation of the drawing]

[0007] [Figure 1] This is an explanatory diagram of an investment environment evaluation system according to an embodiment. [Figure 2] This is an explanatory diagram of the hardware configuration of the embodiment. [Figure 3] This is an explanatory diagram of the processing procedure of the embodiment. [Figure 4] This is an explanatory diagram of the data of the embodiment. [Figure 5] This is an explanatory diagram of a benchmark of an embodiment, where (a) is an explanatory diagram of the case with low resolution and (b) is an explanatory diagram of the case with high resolution. [Modes for carrying out the invention]

[0008] An embodiment of the investment environment evaluation system, investment environment evaluation method, and investment environment evaluation program will be described in accordance with Figures 1 to 5. In this embodiment, it is assumed that the system outputs comments based on confirmed facts regarding market fluctuation factors suitable for explaining fluctuations in the net asset value of a fund (investment trust) managed by the company, according to the analysis period requested by the recipient.

[0009] As shown in Figure 1, the investment environment evaluation system of this embodiment uses a network-connected user terminal 10, an information provision site 15, a support server 20, and an AI system 30.

[0010] (Example hardware configuration) Figure 2 shows an example of the hardware configuration of the information processing device H10, which functions as a user terminal 10, an information provision site 15, a support server 20, an AI system 30, etc.

[0011] The information processing device H10 includes a communication device H11, an input device H12, a display device H13, a storage device H14, and a processor H15. Note that this hardware configuration is an example, and other hardware may be included.

[0012] Communication device H11 is an interface that establishes a communication path with other devices and performs data transmission and reception, such as a network interface or a wireless interface.

[0013] The input device H12 is a device that receives input from users, such as a mouse or keyboard. The display device H13 is a display or touch panel that displays various types of information.

[0014] The storage device H14 is a storage device that stores data and various programs for executing various functions of the user terminal 10, information provision site 15, support server 20, and AI system 30. Examples of storage devices H14 include ROM, RAM, and hard disks.

[0015] Processor H15 controls the respective processes in user terminal 10, information providing site 15, support server 20, and AI system 30 by using the programs and data stored in storage device H14. Examples of processor H15 include, for example, a CPU, an MPU, and the like. This processor H15 expands the program stored in a ROM or the like to a RAM and executes various processes corresponding to various processes. For example, when the application programs of user terminal 10, information providing site 15, support server 20, and AI system 30 are started, processor H15 operates a process that executes the respective processes described later.

[0016] Processor H15 is not limited to performing software processing for all the processes it executes. For example, processor H15 may include a dedicated hardware circuit (for example, an application specific integrated circuit: ASIC) that performs hardware processing for at least a part of the processes it executes. That is, processor H15 may be configured, for example, as follows.

[0017] 〔1〕One or more processors that operate according to a computer program (software) 〔2〕One or more dedicated hardware circuits that execute at least a part of the various processes 〔3〕A circuit (circuitry) including a combination thereof The processor includes a CPU and memories such as a RAM and a ROM, and the memories store program codes or instructions configured to cause the CPU to execute a process. The memory, that is, the computer-readable medium, includes any available medium that can be accessed by a general-purpose or dedicated computer.

[0018] (Functions of each information processing device) Using FIG. 1, the functions of user terminal 10, information providing site 15, support server 20, and AI system 30 used for investment environment evaluation will be described. The user terminal 10 is a computer terminal used by the creator of the report who is a user of this system. Examples of users include fund managers and compliance officers of investment trusts and pension management companies. In this embodiment, the case of creating an operation report required for continuous disclosure in the rules regarding operation reports related to investment trusts and investment corporations will be described, but the report for beneficiaries is not limited to the operation report.

[0019] The information providing site 15 is a computer system for providing various information. The various information includes fundamentals which are indicators and information representing the basic conditions and situations of enterprises and the economy, and information regarding benchmarks which are indicators for comparing operation performance. Enterprise-level fundamentals include information such as profitability (sales, operating profit, etc.), safety (equity ratio, current ratio, etc.), and growth (ROE, etc.).

[0020] Sector / market-level fundamentals include information such as economic indicators (GDP growth rate, unemployment rate, consumer price index, etc.), financial policies (interest rate, exchange rate, etc.), industry trends (technological innovation, deregulation / strengthening of regulations, changes in the competitive environment, etc.), and geopolitical risks. The benchmarks include indicators related to stocks and bonds.

[0021] The support server 20 is a computer system for using generative AI. This support server 20 includes a control unit 21 and a fund information storage unit 22.

[0022] The control unit 21 performs processes (processes including an acquisition stage, an analysis stage, a creation stage, etc.). By executing an investment environment evaluation program for this purpose, the control unit 21 functions as an acquisition unit 211, an analysis unit 212, a creation unit 213, etc.

[0023] The acquisition unit 211 executes a process of acquiring various information from the user terminal 10 and the information providing site 15. The analysis unit 212 performs processing to analyze the status of benchmarks and other factors in the target investment market. The creation unit 213 generates a prompt to be input to the AI ​​system 30 and uses the AI ​​system 30 to execute a process to obtain an operational comment as a response.

[0024] The fund information storage unit 22 stores fund management information for investment trusts (funds). Fund management information is recorded when investment performance is registered. Fund management information includes information such as the fund identifier, investment date, net asset value, total net assets, portfolio, benchmark, and investment comments.

[0025] The fund identifier is an identifier used to identify each fund. The operation date information refers to the information for each calendar day (year, month, and day) during the investment trust period in which this fund was managed.

[0026] The net asset value information is information regarding the price per unit of the investment trust as of this date. The total net asset information refers to the total asset size of the investment trust as of this date. Portfolio information refers to the investment markets in which the entire investment trust's assets are invested, and the proportion of investments in each investment market (asset composition), as of this date.

[0027] Benchmark information refers to benchmarks used to compare the performance of this investment trust. These benchmarks include not only those listed as reference indicators in the cited report and those defined by the investment trust in its terms and conditions or product overview, but also indices that surrogately represent the fluctuations in each market in which the individual investment trust invests. The fund management commentary information contains comments about the investment environment that influenced fluctuations in the net asset value.

[0028] AI system 30 is a computer system that utilizes generative AI technology based on a large-scale language model (LLM) that generates responses to prompts. AI system 30 is configured according to the computational model and parameters, which are elements that determine the structure of the generative AI. In this embodiment, Retrieval-Augmented Generation (RAG), which combines information retrieval and natural language generation, is used as the generative AI technology. This Retrieval-Augmented Generation technology retrieves relevant information in response to a specific question or request and generates an answer based on that information.

[0029] (Support processing) The support process will be explained using Figure 3. First, the control unit 21 of the support server 20 executes the process of acquiring operational performance data (step S11). Specifically, the acquisition unit 211 of the control unit 21 acquires the fund identifier and analysis period of the investment trust to be processed from the user terminal 10. This analysis period can be any period that differs depending on the settlement date of each product, such as a specific period like a weekly or monthly period, or the period for market environment comments used in the investment trust's operational report (from the previous settlement date to the current settlement date). In this case, the acquisition unit 211 acquires the fund management information of the fund to be processed from the fund information storage unit 22.

[0030] As shown in Figure 4, the fund identifier D10 identifies the performance history, the analysis period D11, and the investment target market D12 of the fund management information. Next, the control unit 21 of the support server 20 performs the benchmark identification process (step S12). Specifically, the acquisition unit 211 of the control unit 21 identifies the benchmark recorded in the fund management information to be processed. In this embodiment, there may be cases where investments are made in multiple markets, and furthermore, multiple benchmarks are identified for a single investment target market. As shown in Figure 4, we identify the benchmark D30 corresponding to the target investment market D12.

[0031] Next, the control unit 21 of the support server 20 performs a resolution determination process according to the benchmark (step S13). Specifically, the analysis unit 212 of the control unit 21 calculates the highest and lowest values ​​of the base price during the analysis period in order to determine a threshold according to the range of fluctuation during the analysis period. Next, the analysis unit 212 calculates a threshold by multiplying the range of fluctuation by a predetermined percentage. This threshold determines the resolution according to the investment target market. If the range of fluctuation throughout the analysis period is large, the threshold is increased to increase the criteria for determining the inflection point. On the other hand, if the range of fluctuation throughout the analysis period is small, the threshold is decreased to decrease the criteria for determining the inflection point. Alternatively, the threshold specified by the creator of the operation report may be obtained from the user terminal 10.

[0032] Next, the control unit 21 of the support server 20 identifies the target for processing among multiple benchmarks. Then, it repeats the following process for each benchmark of the target investment market. Here, the control unit 21 of the support server 20 performs a process to identify inflection points using a threshold (step S14). Specifically, the analysis unit 212 of the control unit 21 performs smoothing processing on the temporal change of the base price using a resolution corresponding to the threshold. Next, the analysis unit 212 identifies inflection points where price increases or decreases occur in the smoothed temporal change.

[0033] As shown in Figure 4, the inflection point D50 is identified using benchmark D30 and resolution D40. Figure 5 shows the benchmark fluctuation graph 500, with the benchmark value 501 and the moving average 502. Figure 5(a) shows the fluctuation 505 at low resolution, and Figure 5(b) shows the fluctuation 515 at high resolution. The points where fluctuations 505 and 515 bend are the inflection points.

[0034] Next, the control unit 21 of the support server 20 performs market classification processing (step S15). Specifically, the analysis unit 212 of the control unit 21 determines the market classification at the identified inflection point. Here, if the base price after the inflection point shows a continuous upward or downward trend over a predetermined period, it recognizes the market phase as an "uptrend" or "downtrend," respectively. Also, if the base price fluctuates up and down within a certain range and no clear upward or downward trend is observed, it recognizes the market phase as a "sideways market." The analysis unit 212 then temporarily stores a benchmark, including the market classification, in relation to the resolution. Repeat the above process until all benchmarks have been completed.

[0035] Next, the control unit 21 of the support server 20 performs the creator recognition determination process (step S16). Specifically, the analysis unit 212 of the control unit 21 outputs the market price classification for each resolution to the user terminal 10 and prompts confirmation. If it differs from the creator's recognition, the threshold specified by the creator is obtained from the user terminal 10, and the process of identifying the inflection point (step S14) and subsequent steps is repeated using the threshold. Alternatively, the inflection point may be obtained from the user terminal 10.

[0036] If a resolution matching the creator's perception is obtained from the user terminal 10, the control unit 21 of the support server 20 executes the generation AI input processing (step S17). Specifically, the creation unit 213 of the control unit 21 inputs a prompt to the AI ​​system 30 instructing it to create an operational comment. This prompt includes information on the benchmark, inflection point, market classification, and operational performance (net asset value). The prompt then instructs the AI ​​system to create an operational comment for the benchmark and market classification corresponding to the operational performance.

[0037] In this case, as shown in Figure 4, the AI ​​system 30 uses the target investment market D12 and the inflection point D50 to identify the causal relationship D60 with market fluctuations (market classification). The AI ​​system 30 then identifies benchmarks corresponding to the fluctuation factors in the investment performance. Furthermore, the AI ​​system 30 obtains events that affected the investment trust on the date of the inflection point from the information provision site 15. Here, past investment reports, expert comments, text such as graphs, image data, market time series data, real-time data, etc. can be used. For example, for an "uptrend," "uptrend factors" are extracted, and for a "downtrend," "downtrend factors" are extracted. For a "sideways market," "uptrend factor" events and "downtrend factor" events are extracted. The "uptrend factor" events and "downtrend factor" events mainly extract fundamental factors (mainly medium- to long-term factors) and supply and demand / technical factors (mainly short-term factors). This event includes, at the national and regional levels, GDP growth rate, inflation rate, unemployment rate, political situation, and changes in the decision or direction of monetary policy, and at the corporate level, performance and financial condition. It may also include changes in index constituent stocks, the January effect, pension fund rebalancing, and market participant sentiment. The AI ​​system 30 then predicts the importance D70 of the extracted factors (events) and generates investment environment comments D80. In this case, the AI ​​system 30 is provided with a glossary of terms, such as those standardized by each asset management company, to unify economic terms in comments published externally.

[0038] Next, the control unit 21 of the support server 20 performs the process of identifying the factors causing fluctuations and obtaining comments on the investment environment (step S18). Specifically, the creation unit 213 of the control unit 21 obtains operational comments in response to prompts from the AI ​​system 30.

[0039] Next, the control unit 21 of the support server 20 executes the comment registration process for the investment environment (step S19). Specifically, the creation unit 213 of the control unit 21 records the operation comment in the fund management information of the fund information storage unit 22.

[0040] Next, the control unit 21 of the support server 20 performs the process of identifying comment-related information (step S20). Specifically, the creation unit 213 of the control unit 21 identifies the benchmark included in the operational comment. The creation unit 213 then obtains event information and links as the basis for the benchmark for the analysis period from the information provision site 15.

[0041] Next, the control unit 21 of the support server 20 performs fact-checking (step S21). Specifically, the creation unit 213 of the control unit 21 outputs a confirmation screen to the user terminal 10 to prompt the creator for confirmation. This confirmation screen displays operational comments, event information related to the benchmark for the analysis period, and linked destinations for comparison.

[0042] (Effect of the embodiment) The benchmark market classification is performed according to the resolution, so market classification candidates that match the creator's perception are output.

[0043] (Effects of the embodiment) (1) In this embodiment, the control unit 21 of the support server 20 performs benchmark identification processing (step S12). This makes it possible to identify the benchmark to be used for creating operational comments.

[0044] (2) In this embodiment, the control unit 21 of the support server 20 performs a resolution identification process according to the benchmark (step S13). This makes it possible to calculate a threshold for identifying inflection points and select resolution candidates that correspond to operational performance, even when the investment target market fluctuates according to the situation.

[0045] (3) In this embodiment, the control unit 21 of the support server 20 performs the process of identifying the inflection point (step S14) and the market price classification process (step S15). This makes it possible to classify the market price according to the situation between the inflection points.

[0046] (4) In this embodiment, the control unit 21 of the support server 20 performs recognition and determination processing of the creator (step S16). This makes it possible to correspond to the situation recognition of the creator of the operation report. Therefore, it becomes possible to generate comments on fluctuation factors that are easy to understand intuitively and at the resolution desired by the presenter, corresponding to three types of market fluctuations: "rising market," "sideways market," and "falling market," and are highly convincing.

[0047] (5) In this embodiment, the control unit 21 of the support server 20 performs generation AI input processing (step S17) and processing to identify fluctuation factors and acquire comments on the investment environment (step S18). This makes it possible to acquire comments according to the market classification. Here, by using the latest information on events that may affect investment trusts in RAG, it is possible to reflect information after the learning time and improve the text generation performance of the generation AI.

[0048] (6) In this embodiment, the control unit 21 of the support server 20 performs the identification process of comment-related information (step S20) and the fact-checking process (step S21). This allows for fact-checking by comparing the operational comments obtained from the AI ​​system 30 with market conditions. For example, fact-checking can reduce the workload and risks for reviewers and compliance officers.

[0049] This embodiment can be implemented with the following modifications. This embodiment and the following modifications can be combined with each other to the extent that they do not contradict each other technically. In the above embodiment, a user terminal 10, an information provision site 15, a support server 20, and an AI system 30 are used. The hardware configuration is not limited to these. For example, the information provision site 15 and the AI ​​system 30 may be provided within the support server 20. • The above embodiment describes the case where an investment trust's operational report is prepared as the report, but it is not limited to this. For example, it can be used in the following: • Securities Registration Statement: Disclosure at the time of issuance under the Financial Instruments and Exchange Act • Securities Report: Continuous Disclosure • Prospectus (Terms and Conditions): Rules concerning the preparation of the prospectus. • Prospectus: Disclosure at the time of issuance and continuous updates as stipulated in the Financial Services Agency's supervisory guidelines and the association's guidelines published from time to time. • Full version: Continuous disclosure based on Financial Services Agency supervisory guidelines and association guidelines published as needed. • Monthly report: Continuous disclosure under the rules concerning investment trust and investment corporation operation reports, etc.

[0050] In the above embodiment, the control unit 21 of the support server 20 performs benchmark identification processing (step S12). Here, a predetermined benchmark recorded in the fund information storage unit 22 is used. The benchmark is not limited to a predetermined one, as long as it is information that may have an impact on operations. For example, a benchmark specified by the person who prepared the operation report may be obtained from the user terminal 10. Alternatively, a benchmark used in past operation reports of the same type of fund may be used.

[0051] In the above embodiment, the control unit 21 of the support server 20 performs benchmark identification processing (step S12). In addition, a synthetic benchmark may be used. A synthetic benchmark is a benchmark created by combining multiple financial instruments or indicators, rather than using actual market indices or individual stocks.

[0052] In the above embodiment, the control unit 21 of the support server 20 performs a resolution identification process according to the benchmark (step S13). Here, multiple thresholds may be set for a single benchmark to identify inflection points at different resolutions.

[0053] Furthermore, in the above embodiment, the threshold is determined according to the range of fluctuations during the analysis period, but the method for determining the threshold is not limited to the range of fluctuations. For example, in addition to "%" and "bps", it may also be determined by the trading unit for each market, such as "yen".

[0054] In the above embodiment, the control unit 21 of the support server 20 performs a recognition determination process of the creator (step S16) before the generation AI input processing (step S17). Alternatively, the agreement with the creator's recognition may be determined in the fact-checking process (step S21). In this case, the creator confirms the recognition, including operational comments.

[0055] In the above embodiment, for investment trusts managed by the company, comments based on confirmed facts regarding market fluctuation factors suitable for explaining fluctuations in net asset value are output, according to the analysis period requested by the recipient. The scope of application of this disclosure is not limited to investment trusts managed by the company. For example, for investment trusts held by customers, comments based on confirmed facts regarding market fluctuation factors suitable for explaining fluctuations in net asset value are output, according to the investment period required for after-sales service. This makes it possible to provide appropriate explanations of market fluctuations corresponding to the investment trusts held by customers. Furthermore, while after-sales service is normally provided on a monthly basis, it is possible to respond even if after-sales service becomes necessary on a weekly basis due to rapid market fluctuations or other reasons. [Explanation of symbols]

[0056] 10...User terminal, 15...Information provision site, 20...Support server, 21...Control unit, 211...Acquisition unit, 212...Analysis unit, 213...Creation unit, 22...Fund information storage unit, 30...AI system.

Claims

1. An investment environment evaluation system comprising a user terminal and a control unit connected to a generating AI, The control unit, From the aforementioned user terminal, information regarding the fund and the period to be analyzed is obtained. Obtain benchmarks related to the investment markets of the aforementioned fund, With respect to the benchmark, the inflection points of the analysis period are identified according to a resolution appropriate for the analysis period, and the market phases between each identified inflection point are classified. Regarding the aforementioned market phase, comments on the fund's investment environment are obtained from the aforementioned generating AI. An investment environment evaluation system characterized by outputting the aforementioned comments and the basis for those comments to the user terminal.

2. The investment environment evaluation system according to claim 1, characterized in that the control unit acquires the benchmark corresponding to the investment target market of the fund.

3. The investment environment evaluation system according to claim 1 or 2, characterized in that the control unit identifies the inflection point with the resolution corresponding to market fluctuations during the analysis period of the fund.

4. The investment environment evaluation system according to claim 1 or 2, characterized in that the control unit acquires the inflection point from the user terminal.

5. The investment environment evaluation system according to claim 1, characterized in that the control unit acquires events that occurred during the analysis period and affected the fund as factors causing fluctuations.

6. A method for evaluating the investment environment using an investment environment evaluation system that includes a user terminal and a control unit connected to a generating AI, The control unit, From the aforementioned user terminal, information regarding the fund and the period to be analyzed is obtained. Obtain benchmarks related to the investment markets of the aforementioned fund, With respect to the benchmark, the inflection points of the analysis period are identified according to a resolution appropriate for the analysis period, and the market phases between each identified inflection point are classified. Regarding the aforementioned market phase, comments on the fund's investment environment are obtained from the aforementioned generating AI. An investment environment evaluation method characterized by outputting the aforementioned comments and the basis for those comments to the user terminal.

7. A program for evaluating the investment environment using an investment environment evaluation system that includes a user terminal and a control unit connected to a generating AI, The control unit, From the aforementioned user terminal, information regarding the fund and the period to be analyzed is obtained. Obtain benchmarks related to the investment markets of the aforementioned fund, With respect to the benchmark, the inflection points of the analysis period are identified according to a resolution appropriate for the analysis period, and the market phases between each identified inflection point are classified. Regarding the aforementioned market phase, comments on the fund's investment environment are obtained from the aforementioned generating AI. An investment environment evaluation program that functions as a means of outputting the aforementioned comments and the basis for those comments to the user terminal.

Images