Intelligent financial asset profit and loss monitoring method

By employing data governance and intelligent evaluation methods, and utilizing algorithms such as Pattern Finder to screen and monitor financial asset data, the problem of insufficient intelligence in quantitative evaluation in existing technologies has been solved. This enables efficient and accurate monitoring of financial asset gains and losses, adapts to explosive data growth and multi-dimensional analysis, and reduces labor costs.

CN118946903APending Publication Date: 2024-11-12徐维
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202380017556.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-11-08
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

Existing technologies lack quantitative assessment of indicators such as the time sequence, application scenarios, time dependencies, and asset attributes of financial assets using methods such as pattern finder, value distribution, value matcher, and boolean analyzer, resulting in insufficient intelligence and accuracy in monitoring the profit and loss of financial assets.

Method used

Data governance techniques are employed, including pattern finder, value distribution, and boolean analyzer for data range selection, CN2 rule classification algorithm and KNN classification algorithm for data verification, deep learning and prediction threshold setting for screening and monitoring of financial asset data, metadata range and business logic for processing and tracing, PREDICTIONS algorithm and CO NFUSION MATRIX algorithm for comprehensive evaluation, and ROC analysis to ensure the accuracy of monitoring results.

Benefits of technology

It achieves intelligent and efficient monitoring of the gains and losses of financial assets, reduces the impact of dirty and heterogeneous data, improves monitoring efficiency and accuracy, supports multi-dimensional analysis and rapid decision-making, has flexibility and adaptability, and reduces the cost of manual investigation.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

An intelligent financial asset profit and loss monitoring method is a financial asset profit and loss calculation and automatic monitoring method based on big data prediction, an evaluation algorithm, coverage service logic limitation and underlying asset data, and comprises the following steps: establishing technical standards among multiple systems, and completing data set concentration through preset codes; the method comprises the following steps of: intelligently judging and selecting data through algorithms such as a value distribution algorithm, a value matcher, a boolean angle, a pattern finder and the like, and then executing a specific monitoring algorithm. The method comprises the following steps of: carrying out intelligent judgment and selection on the data through the algorithms such as a value distribution algorithm, a value matcher, a boolean angle, a pattern finder and the like;
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the fields of financial asset evaluation and big data profit and loss monitoring, and in particular to an intelligent financial asset profit and loss monitoring method. Background Art

[0002] Data governance technology. In the digital age, while the value of data from various dimensions of financial asset management for asset profit and loss monitoring continues to increase, dirty data and heterogeneous data issues are becoming increasingly prominent. The development of data governance technology has created conditions for intelligent and personalized monitoring of financial asset profit and loss. SUMMARY OF THE INVENTION Technical issues

[0003] Financial asset profit and loss monitoring generally adopts two methods: qualitative evaluation and quantitative evaluation. In the currently known technical field, there is no systematic quantitative evaluation method that uses pattern finder, value distribution, value matcher, booleananalyzer and other methods to identify the timing, application scenarios, time (timeliness) dependency, asset attributes, categories and other indicators of financial asset business and then implement data range selection based on prediction and evaluation algorithms.

[0004] The present invention provides an intelligent financial asset profit and loss monitoring method, in which data governance and data quality judgment mechanisms are used, including: quantitative evaluation of data sets through methods such as pattern finder, value distribution, value matcher, boolean analyzer, etc. This link will select the data range included in the calculation, which is the basic link.

[0005] At the beginning of implementation, the data values ​​generated by exhaustive enumeration are secondary verified and confidence predicted by using CN2 rule classification algorithm and KNN classification algorithm. Based on the prediction threshold setting, the screened data elements are selected for monitoring and calculation.

[0006] In the implementation process, prediction, evaluation algorithms and deep learning are applied, covering business logic limitations and data metadata. This includes: through the disposal and inspection of the preset codes of financial assets, the centralized registration of financial asset data sets is completed, and the various types of financial asset data identified and transmitted are processed according to the preset rules for screening, judgment and tracing.

[0007] When further implementing data indicator processing, the metadata scope shall at least include asset portfolio code, asset portfolio abbreviation, asset portfolio value date, asset portfolio maturity date, asset portfolio size, benchmark lower limit, benchmark upper limit, asset portfolio net value, asset portfolio expected net value (based on performance benchmark), net value deviation, annualized since establishment / annualized for this cycle, change in net value compared with the previous day, annualized in the past month, annualized in the past three months, excess performance reward ratio, fixed expense ratio, realized asset-side income, fixed expenses, equity asset size (after penetration), proportion of equity-inclusive asset account size, equity proportion, direct financing, debt financing, interbank loans, corporate bonds, local enterprise bonds, bond asset management plans, medium- and short-term bond funds, long-term bond funds, primary bond funds, secondary bond funds, mixed funds, stock funds, mixed asset management plans, money market funds, demand deposits, the lower limit of future asset yield required by the benchmark, and the upper limit of future asset yield required by the benchmark.

[0008] The lower limit of the future asset yield required by the performance benchmark, for open-ended products, if [(product scale * performance benchmark lower limit / 365 * (product maturity date - product value date + 1) * net value at the beginning of this week + fixed fees - realized asset-side income] > 0, [(product scale * performance benchmark lower limit / 365 * (product maturity date - product value date + 1) * net value at the beginning of this week + fixed fees - realized asset-side income] / product scale / product net value / (product maturity date - business date) * 365, otherwise = 0. For closed-end products, if [(product scale * performance benchmark lower limit / 365 * (product maturity date - product value date + 1) * 1 + fixed fees - realized asset-side income] > 0, product scale * performance benchmark lower limit / 365 * (product maturity date - product value date + 1) * 1 + fixed fees - realized asset-side income / product scale / product net value / (product maturity date - business date) * 365, otherwise = 0.

[0009] The upper limit of future asset yield required by the performance benchmark, for open-ended products, if [(product scale * upper limit of performance benchmark / 365 * (product maturity date - product value date + 1) * net value at the beginning of this week + fixed fees - realized asset-side income]>0, [(product scale * lower limit of performance benchmark / 365 * (product maturity date - product value date + 1) * net value at the beginning of this week + fixed fees - realized asset-side income] / product scale / product net value / (product maturity date - business date) * 365, otherwise = 0. For closed-end products, if [(product scale * lower limit of performance benchmark / 365 * (product maturity date - product value date + 1) * 1 + fixed fees - realized asset-side income]>0, product scale * upper limit of performance benchmark / 365 * (product maturity date - product value date + 1) * 1 + fixed fees - realized asset-side income / product scale / product net value / (product maturity date - business date) * 365, otherwise = 0.

[0010] When the market interest rate rises by 10BP, the change in product net value = weighted duration of bonds held * market value ratio of bonds held * 10BP / 100, weighted duration of bonds held = ∑ duration of bond i * market value ratio of bond i.

[0011] Calculate the daily returns for the past 12 months, and then arrange them in descending order. The data ranked 95% (i.e. the bottom 5%) is the VaR value at a 95% confidence level. Automatically verify the comprehensiveness of asset data processing. The asset data processed in the previous step enters a specific sandbox for analysis and observation. Assist the PREDICTIONS algorithm and CONFUSION MATRIX algorithm to conduct a systematic and comprehensive evaluation of the asset data satisfaction, and the evaluation results are presented in a graded and classified manner. Further verify the timeliness of asset data processing, monitor metadata changes and asset data report updates in real time, and respond to changes in a timely manner. Verify data screening, conduct secondary screening and verification of the asset metadata screened in the previous step, and use other conventional methods or qualitative methods to assist in evaluating the profit and loss monitoring results.

[0012] Further monitoring and evaluation to achieve a closed-loop management: Regularly monitor and evaluate the results of asset profit and loss analysis to ensure accuracy. The core method is problem discovery tracking, assisted by ROC (receiver operating characteristic curve). ROC analysis involves the drawing of univariate and multivariate ROC and PR curves for asset metadata effectiveness evaluation, determination of AUC and optimal cut-off values, and finding the optimal cut-off value through curve sensitivity and specificity.

[0013] The beneficial effect of the present invention is that the application of the intelligent financial asset profit and loss monitoring method of the present invention can reduce dirty data and heterogeneous effects, reduce the cost of manual investigation, and realize intelligent and efficient financial asset profit and loss indicator monitoring. In particular, the use of this method can adapt to the explosive growth of data, and has plasticity and attribution, and is applicable to different depths, breadths and multi-dimensional analyses involved in asset profit and loss analysis. It can improve the efficiency and accuracy of financial asset profit and loss monitoring, and support faster and more accurate financial asset profit and loss analysis and investment decisions.

[0014] Access external data, such as Wind, Caihui Dazhihui, Flush, China Bond Registration Center, China Securities Registration Center, and Financial Management Registration Center data.

[0015] Relying on PYTHON, ORANGE and other data analysis and mining tools, the key points of the technical solution in the method of the present invention are implemented by writing an execution program, intelligently identifying and screening internal and external financial asset related data, and executing monitoring indicator processing. It can also further rely on the visualization function of various mainstream report plug-ins to realize the visualization of monitoring indicators, and rely on email access and SMS platform to realize the sending of financial asset monitoring and early warning information.

[0016] The SPPI test (contractual cash flow test of financial assets) can be used to verify the accuracy of the data under the method of the present invention.

[0017] Access external data, prepare the company's own financial asset data, and the overall scope of metadata meets the minimum requirements of the present invention.

[0018] Although the present invention has been described according to a limited number of embodiments, it will be apparent to those skilled in the art, with the benefit of the above description, that other embodiments may be envisioned within the scope of the invention thus described. In addition, it should be noted that the language used in this specification is selected primarily for readability and teaching purposes, rather than for explaining or defining the subject matter of the present invention. Therefore, many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the appended claims. The disclosure of the present invention is illustrative, not restrictive, with respect to the scope of the present invention, which is defined by the appended claims.

Claims

1. An intelligent financial asset profit and loss monitoring method, characterized in that: Apply data governance ideas and data quality problem determination mechanisms, including: quantitative assessment of the quality of financial asset data sets through methods such as pattern finder, value distribution, value matcher, boolean analyzer, etc. This link will select the data range included in the calculation, which is the basic link.

2. An intelligent financial asset profit and loss monitoring method as claimed in claim 1, characterized in that: Also includes: Through exhaustive enumeration and the use of CN2 rule classification algorithm and KNN classification algorithm, the data values ​​generated by exhaustive enumeration are secondary verified and confidence predicted. Based on the prediction threshold setting, the asset data after intelligent judgment is selected for specific monitoring indicator calculation and judgment.

3. The intelligent financial asset profit and loss monitoring method according to claim 2, characterized in that: Also includes: The metadata extraction scope shall at least include the asset portfolio code, asset portfolio abbreviation, asset portfolio value date, asset portfolio maturity date, asset portfolio size, benchmark lower limit, benchmark upper limit, asset portfolio net value, asset portfolio expected net value (based on performance benchmark), net value deviation, annualized since inception / annualized for this cycle, change in net value compared with the previous day, annualized for the past 1 month, annualized for the past 3 months, excess performance reward ratio, fixed expense ratio, realized asset-side income, fixed expense, equity asset size (after penetration), proportion of equity-inclusive asset account size, equity ratio, direct financing, debt financing, interbank borrowing, corporate bonds + local enterprise bonds, bond asset management plan, medium- and short-term bond funds, long-term bond funds, primary bond funds / secondary bond funds, mixed funds, stock funds, mixed asset management plans, money market funds, demand deposits, the lower limit of future asset yield required for the benchmark, and the upper limit of future asset yield required for the benchmark.