[go: up one dir, main page]

CN118885912B - Attribution analysis method and attribution analysis system applied to complex indexes - Google Patents

Attribution analysis method and attribution analysis system applied to complex indexes Download PDF

Info

Publication number
CN118885912B
CN118885912B CN202411346425.8A CN202411346425A CN118885912B CN 118885912 B CN118885912 B CN 118885912B CN 202411346425 A CN202411346425 A CN 202411346425A CN 118885912 B CN118885912 B CN 118885912B
Authority
CN
China
Prior art keywords
indicator
indicator components
components
index
complex
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202411346425.8A
Other languages
Chinese (zh)
Other versions
CN118885912A (en
Inventor
韩统优
刘子昕
张镓奇
陈超
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Bank Of Shanghai Co ltd
Original Assignee
Bank Of Shanghai Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Bank Of Shanghai Co ltd filed Critical Bank Of Shanghai Co ltd
Priority to CN202411346425.8A priority Critical patent/CN118885912B/en
Publication of CN118885912A publication Critical patent/CN118885912A/en
Application granted granted Critical
Publication of CN118885912B publication Critical patent/CN118885912B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/211—Selection of the most significant subset of features
    • G06F18/2113—Selection of the most significant subset of features by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses an attribution analysis method and system applied to complex indexes, which relate to the technical field of business data processing and comprise the steps of comparing related business link flows with a plurality of index components, constructing mapping relations between the related business link flows and the index components, screening and selecting the index components through redundancy to reserve part of the index components, determining interaction relations between the index components according to the association relations between the index components, determining whether coupling effects exist in the interaction relations between the index components, analyzing and quantifying the coupling effects if the coupling effects exist, and visually displaying the index components, the interaction relations and the coupling effects based on the mapping relations between the related business link flows and the index components, so that attribution display of business links is realized. The adaptability and the accuracy of the attribution analysis of the complex indexes are improved, and the traceability of the attribution analysis is ensured.

Description

Attribution analysis method and attribution analysis system applied to complex indexes
Technical Field
The invention relates to the technical field of business data processing, in particular to an attribution analysis method and system applied to complex indexes.
Background
In the field of complex and changeable commercial finance, due to an analysis scheme, key factors influencing business results are accurately identified by deeply tracking and analyzing multidimensional and multi-level data indexes by means of advanced background technology. The technology not only covers the fine disassembly of the atomic indexes and the composite indexes, but also combines various algorithms such as time sequence analysis and the like, ensures that the root cause of the problem can be rapidly positioned in the complex business decision process, and assists the management layer to realize the intelligent decision of data driving.
In the prior art, only reasons of complex indexes are often given, complex effects among the reasons are not analyzed, and adaptability and accuracy of attribution analysis are poor.
Therefore, how to improve the adaptability and accuracy of the complex index attribution analysis is a problem which needs to be solved at present.
Disclosure of Invention
The invention aims to solve the problems of poor adaptability and low accuracy of complex index attribution analysis in the prior art, and provides an attribution analysis method applied to complex indexes, which comprises the following steps of,
Acquiring a complex index model and a business target aimed at by a complex index in the financial or business field, obtaining a related business link flow through mapping the business target aimed at by the complex index, disassembling the complex index model according to a dimension level to obtain a plurality of index components, comparing the related business link flow with the plurality of index components to supplement the index components and construct a mapping relation between the related business link flow and the index components;
Analyzing the redundancy among the index components, screening and selecting the index components through the redundancy to reserve part of the index components, dividing the types of the index components according to the relation between the complex index and each index component, associating the index components according to the types of the index components, and determining the association relation among the index components;
determining interaction relation among index components according to the association relation among the index components, determining whether a coupling effect exists in the interaction relation among the index components, and analyzing and quantifying the coupling effect if the coupling effect exists;
and visually displaying the index components, the interaction relation and the coupling effect based on the mapping relation between the related business link flow and the index components, thereby realizing attribution display of the business links.
In some embodiments of the present application, a complex index model is disassembled according to a dimension hierarchy to obtain a plurality of index components, including,
Determining a function of a complex index according to a complex index model, decomposing the function of the complex index into a plurality of variables, confirming the dimension level to which each variable belongs, searching a plurality of other variables close to the variable under the same dimension level, and calculating the stability of each variable and each variable in the plurality of other variables close to the variable;
The judging period of each variable is distributed through the attribute of the variable, the variation coefficient of the variable under each judging period is integrated to determine the stability of each variable, and the numerical value of the variable and the stability thereof are used as index components.
In some embodiments of the application, redundancy between index components is analyzed, including,
Constructing a variable curve graph of the index component variable which changes along with time according to the numerical value of the index component variable, determining the text keyword of each index component variable, and comparing the similarity of the text keywords of the variables among the index components so as to determine the candidate index components with similar text relations;
Comparing variable curve graphs among the index components to be selected with similar text relations to obtain variable curve similarity, and calculating redundancy among the index components to be selected with similar text relations according to the variable curve similarity of the index components to be selected with similar text relations, the similarity of text keywords and the stability of variables; Wherein, the method comprises the steps of, For redundancy between candidate pointer components having similar textual relationships,For redundant weights of the variate similarity of the candidate indicator components with similar text relationships,For the variate similarity of the candidate pointer components with similar text relationships,For redundant weights of the similarity of text keywords of candidate indicator components having similar text relationships,For the similarity of text keywords of the candidate indicator components having similar text relationships,For redundant weights of stability of variables of the candidate pointer component having similar textual relationships,In order to convert the coefficients of the coefficients,、Stability of variables for the candidate pointer components having similar textual relationships.
In some embodiments of the present application, the index components are filtered by redundancy to preserve a portion of the index components, including,
If the redundancy exceeds the redundancy threshold, removing the index component to be selected with lower stability of the variable among the index components to be selected with similar text relationship, and reserving the index component to be selected with higher stability of the variable;
otherwise, all the index components to be selected in the index components to be selected with similar text relations are reserved.
In some embodiments of the present application, the types of index components are further partitioned according to relationships between the complex index and each index component, including,
The types of the index components comprise a direct index component, a semi-direct index component and an indirect index component;
Constructing a fish bone map of complex indexes according to related business link flows and index components, thereby determining fish heads and fish bones in the fish bone map, wherein the fish heads are complex indexes, the fish bones are index components, the fish bone map is converted into index components to influence the path map, the index components are nodes in the path map, and the complex indexes are path end points;
Calculating the correlation between two adjacent nodes with a communication relation, defining the path length between the two nodes through the correlation, dividing index components of the nodes directly communicated with the path end point into direct index components, calculating the path length from the nodes to the path end point, dividing the index components of the nodes with the path length within a preset length range into semi-direct index components, and determining the route from the nodes to the path end point;
the index components of the other nodes are divided into indirect index components and the route of the node to the path end point is determined.
In some embodiments of the present application, the index components are associated according to the type of the index component, and the association relationship between the index components is determined, including,
Corresponding index components are associated by direct index components, corresponding index components are associated by path routes related to semi-direct index components, and corresponding index components are associated by path routes related to indirect index components.
In some embodiments of the application, and determining whether there is a coupling effect in the interaction relationship between the index components, includes,
Carrying out standardization processing on a plurality of variables involved in the interaction relation among index components, calculating covariance matrixes among the plurality of variables, solving the covariance matrixes to obtain characteristic values and characteristic vectors, obtaining a plurality of new variables, giving weights to the new variables through the characteristic values, and carrying out weighted summation on the new variables to obtain an initial comprehensive representative index;
Adjusting an initial comprehensive representative index through the stability of a plurality of variables involved in the interaction relation between index components to obtain a comprehensive representative index; Wherein, the method comprises the steps of, In order to synthesize the representative index of the present invention,For the purpose of initially synthesizing the representative index,To indicate the number of index members that are in an interaction relationship,To the first of the interaction relationshipStability of the individual index assembly,For a maximum stability of the index assembly in which an interaction relationship exists,The factor of the fluctuation is used to determine,For the fluctuation factor mapped from the stability maximum of the index components having an interaction relationship,Calculating the nonlinear dependency degree between the comprehensive representative index and the complex index, and judging whether the coupling effect exists or not according to the nonlinear dependency degree.
In some embodiments of the application, the method further comprises,
And if the coupling effect does not exist, visually displaying the index components and the interaction relation based on the mapping relation between the related business link flow and the index components.
Correspondingly, the application also provides a attribution analysis system applied to the complex index, which comprises,
The system comprises a first module, a second module and a third module, wherein the first module is used for acquiring a complex index model and a business target aimed at by a complex index in the financial or business field, obtaining a relevant business link flow through mapping the business target aimed at by the complex index, disassembling the complex index model according to a dimension level to obtain a plurality of index components, and comparing the relevant business link flow with the plurality of index components so as to supplement the index components and construct a mapping relation between the relevant business link flow and the index components;
The second module is used for analyzing the redundancy among the index components, screening and selecting the index components through the redundancy to reserve part of the index components, dividing the types of the index components according to the relation between the complex index and each index component, associating the index components according to the types of the index components, and determining the association relation among the index components;
A third module, configured to determine an interaction relationship between the index components according to the association relationship between the index components, determine whether a coupling effect exists in the interaction relationship between the index components, and if the coupling effect exists, analyze and quantify the coupling effect;
and the fourth module is used for visually displaying the index components, the interaction relation and the coupling effect based on the mapping relation between the related business link flow and the index components, so that attribution display of the business links is realized.
Compared with the prior art, the invention has the beneficial effects that:
1. the related business link flow and the index components are compared, the supplement of the index components is completed, and the mapping relation between the index components is constructed, so that the subsequent comparison is convenient.
2. The index components are screened and selected through redundancy to keep part of the index components, redundant index components with similar meaning are removed, the independence of the index components is guaranteed, and the reliability of analysis of the index components is improved.
3. The interaction relation among the index components is determined according to the association relation among the index components, whether the coupling effect exists in the interaction relation among the index components is determined, the association among the index components is accurately given, the interaction relation among the index components is obtained, whether the coupling effect is generated by the interaction relation is analyzed, adaptability and accuracy of complex index attribution analysis are improved, and traceability of attribution analysis is guaranteed.
Drawings
FIG. 1 is a flow chart of an attribution analysis method applied to complex indexes;
Fig. 2 is a schematic structural diagram of an attribution analysis system applied to complex indexes according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.
Referring to fig. 1, a attribution analysis method applied to complex indexes includes the steps of:
Step S101, a complex index model and a business target aimed at by complex indexes in the financial or business field are obtained, a relevant business link flow is obtained through mapping of the business target aimed at by the complex indexes, the complex index model is disassembled according to a dimension level to obtain a plurality of index components, and the index components are supplemented and a mapping relation between the relevant business link flow and the index components is constructed according to comparison of the relevant business link flow and the plurality of index components.
In this embodiment, the complex index model (i.e., the functional relationship of the complex index) refers to a multi-dimensional and multi-level data index system used for reflecting business performance, operation status, and the like in the financial or business field. These models typically contain a number of variables and parameters that can comprehensively reflect aspects of a particular business objective or performance. For example, a comprehensive profitability indicator is a complex indicator that measures the comprehensive profitability of a financial institution over a period of time that comprehensively accounts for the performance of multiple aspects of revenue, cost control, asset quality, capital sufficiency, and the like. The business link flow is as follows:
1. product design and development (market share growth rate after new product is marketed, product gross interest rate, etc.):
and designing a competitive financial product which meets the market demand.
Optimizing the product pricing strategy and ensuring profit space.
2. Marketing and sales (marketing fee input-output ratio, new customer acquisition cost, customer repurchase rate, etc.):
an effective marketing plan is formulated, and the brand awareness degree and the product permeability are improved.
And the sales channel is expanded, and the sales volume is increased.
3. Risk management (bad loan rate, capital sufficiency, risk weighted property duty, etc.):
And a sound risk management system is established, so that credit risks, market risks and operation risks are effectively controlled.
And (5) carrying out risk assessment regularly and adjusting a risk management strategy in time.
4. Cost control (operation cost accounts for total income proportion, average yield value, etc.):
And the operation flow is optimized, and the operation cost is reduced.
And the budget management is enhanced, and various expenses are ensured to be reasonable and efficient.
5. Customer service and relationship management (customer satisfaction survey score, customer complaint treatment timeliness, etc.):
Providing high-quality customer service and enhancing customer viscosity.
Through the customer relationship management system, customer requirements are deeply known, and customized services are provided.
In this embodiment, the design of the complex index is initially to better reflect and track the implementation of the business objective. Thus, there is a close relationship between these metrics and business objectives. By further analyzing complex metrics, we can more clearly understand how they correspond to business objectives and how they help achieve business objectives. The complex index model is disassembled according to a dimension level, the dimension level comprises time, space, user characteristics, product attributes, channel sources and the like, and related business link flows and a plurality of index components are calibrated, so that the index components are supplemented, and related indexes can be supplemented, so that the integrity of the indexes is ensured. The mapping relation between the related business link flow and the index components is that different index components correspond to different business links, and a certain part of business conditions are reflected.
In some embodiments of the present application, a complex index model is disassembled according to a dimension hierarchy to obtain a plurality of index components, including,
Determining a function of a complex index according to a complex index model, decomposing the function of the complex index into a plurality of variables, confirming the dimension level to which each variable belongs, searching a plurality of other variables close to the variable under the same dimension level, and calculating the stability of each variable and each variable in the plurality of other variables close to the variable;
The judging period of each variable is distributed through the attribute of the variable, the variation coefficient of the variable under each judging period is integrated to determine the stability of each variable, and the numerical value of the variable and the stability thereof are used as index components.
In this embodiment, the complex index obtained by the complex index model is generally formed by combining a plurality of sub-indexes or variables through a certain mathematical relationship, and is used for measuring the achievement of a specific business objective, and it is necessary to identify the mathematical relationship of each sub-index to form the complex index, that is, the functional expression of the complex index. This function may be a linear combination, polynomial, logarithmic, exponential function, or a complex form of these functions. For example, the complex index Y may be formed by combining a plurality of variables of X1, X2,., xn, etc. by a certain functional relationship f, expressed as y=f (X1, X2,., xn). Decomposing the function into a plurality of variables, for example, if the function is a polynomial, it is necessary to identify the coefficients of each term and the corresponding variables.
In this embodiment, the index is extended by searching for several other variables that are close to the variable (possibly representing similar meaning, representing related meaning) in the same dimension level, and by using correlation analysis or cluster analysis to find variables that are closely related to each other in the same dimension. Different variables possibly have different fluctuation conditions, a judging period is allocated according to the business meaning of the variable, the variation coefficient under each period is calculated, the variation coefficients under a plurality of periods are integrated (weighted summation or weighted averaging) to obtain an overall variation coefficient, and the overall variation coefficient corresponds to the stability of one variable. The index component includes two parts, a variable and an attribute of its stability.
For example, when the complex index is a bank credit card business profit index, the bank credit card business profit index function relationship (such as the independent variables of interest income, annual fee income, commission income, bad account loss and operation cost of credit card business) is as follows: Wherein B is a credit card business profit index of the bank, Is a constant term, typically representing the base net profit without any arguments (i.e., without interest income, annual fee income, commission income, bad account loss and operating cost),Coefficients representing interest revenue, annual fee revenue, commission revenue, bad account loss and operation cost of the credit card business,Representing interest revenue, annual fee revenue, commission revenue, bad account loss and operation cost of the credit card business,Is the tax rate of tax, and is used to calculate the tax that should be paid in advance of the profit (i.e., the profit without considering the influence of tax). Pre-tax profit is obtained by subtracting bad account loss and operating costs from interest revenue, annual fee revenue, commission revenue. The tax rate is multiplied by the pre-tax profit (the portion after deducting bad account loss and operating costs) to obtain the tax due, which is then deducted from the pre-tax profit to obtain the final net profit. The decision period is allocated according to the business meaning of the variables, specifically, for example, the interest income mainly comes from the interest fee generated by the non-on-time repayment of the cardholder, and the repayment behavior of the cardholder may be affected by various factors, such as income level, consumption habit and the like, because the interest income is closely related to the repayment behavior of the cardholder, and the factors are not usually changed very frequently. Thus, a relatively long period may be selected to observe the stability of interest revenue, such as a monthly or quarterly period. The commission revenues include a pick-up commission, a staging commission, and the like, which are typically associated with a particular transaction activity of the cardholder. The fluctuation in commission revenue may be more frequent because it is directly related to the cardholder's transaction frequency and amount. In order to capture the trend of the change more accurately, a shorter determination period, such as a month or a week, may be selected.
It should be noted that, when the complex index is a credit card business profit index of a bank, the variables of the complex index may be further divided, or the variables included in other complex indexes are more, and for reasons of space, the complex indexes are not displayed one by one.
Step S102, analyzing the redundancy among the index components, screening and selecting the index components through the redundancy to reserve part of the index components, dividing the types of the index components according to the relation between the complex index and each index component, associating the index components according to the types of the index components, and determining the association relation among the index components.
In this embodiment, redundancy among index components means that different indexes may describe the same or very similar condition, repetition may occur, resulting in lower accuracy of attribution analysis, where redundancy is defined and index components are screened.
In some embodiments of the application, redundancy between index components is analyzed, including,
Constructing a variable curve graph of the index component variable which changes along with time according to the numerical value of the index component variable, determining the text keyword of each index component variable, and comparing the similarity of the text keywords of the variables among the index components so as to determine the candidate index components with similar text relations;
Comparing variable curve graphs among the index components to be selected with similar text relations to obtain variable curve similarity, and calculating redundancy among the index components to be selected with similar text relations according to the variable curve similarity of the index components to be selected with similar text relations, the similarity of text keywords and the stability of variables; Wherein, the method comprises the steps of, For redundancy between candidate pointer components having similar textual relationships,For redundant weights of the variate similarity of the candidate indicator components with similar text relationships,For the variate similarity of the candidate pointer components with similar text relationships,For redundant weights of the similarity of text keywords of candidate indicator components having similar text relationships,For the similarity of text keywords of the candidate indicator components having similar text relationships,For redundant weights of stability of variables of the candidate pointer component having similar textual relationships,In order to convert the coefficients of the coefficients,、Stability of variables for the candidate pointer components having similar textual relationships.
In this embodiment, the similarity of the text keywords of the variables between the index components is compared, and the similarity of the text keywords (the similarity of the meaning of the variables) is defined by the similarity or the proximity of the keywords. The stability may be similar among similar index components, so the stability similarity is determined by considering the stability deviation.
In this embodiment, the description text of the index component variables may be subjected to word segmentation processing using an NLP tool, TF-IDF (word frequency-inverse document frequency) calculates TF-IDF values for each vocabulary, TF is word frequency, IDF is inverse document frequency (here, "document" may be regarded as a description text set of all the index component variables), and TF-IDF (word frequency-inverse document frequency) weighting method may be used to calculate similarity between keyword sets.
It should be noted that, generally, there are two index components to be selected having similar text relationships, so the stability in the redundancy formula is the difference between two parameters, and if there are more than two index components to be selected having similar text relationships, the redundancy formula can be correspondingly modified to be the difference between multiple parameters.
In some embodiments of the present application, the index components are filtered by redundancy to preserve a portion of the index components, including,
If the redundancy exceeds the redundancy threshold, removing the index component to be selected with lower stability of the variable among the index components to be selected with similar text relationship, and reserving the index component to be selected with higher stability of the variable;
otherwise, all the index components to be selected in the index components to be selected with similar text relations are reserved.
In this embodiment, when the redundancy is higher, when the multiple index components to be selected are screened and retained, the index component corresponding to the party with the highest stability is retained, so that the stability of the index component is improved.
In some embodiments of the present application, the types of index components are further partitioned according to relationships between the complex index and each index component, including,
The types of the index components comprise a direct index component, a semi-direct index component and an indirect index component;
Constructing a fish bone map of complex indexes according to related business link flows and index components, thereby determining fish heads and fish bones in the fish bone map, wherein the fish heads are complex indexes, the fish bones are index components, the fish bone map is converted into index components to influence the path map, the index components are nodes in the path map, and the complex indexes are path end points;
Calculating the correlation between two adjacent nodes with a communication relation, defining the path length between the two nodes through the correlation, dividing index components of the nodes directly communicated with the path end point into direct index components, calculating the path length from the nodes to the path end point, dividing the index components of the nodes with the path length within a preset length range into semi-direct index components, and determining the route from the nodes to the path end point;
the index components of the other nodes are divided into indirect index components and the route of the node to the path end point is determined.
In this embodiment, the path length between two nodes is defined by the correlation, and if the correlation is high, the path length is short. And helping to construct a fishbone diagram of the complex index according to the related business link flow and index components, and analyzing the influence path of the index components of the complex index. The direct index component is a variable that directly affects the complex index, the semi-direct index component is a variable that affects the complex index by affecting other direct indexes, and has a larger influence (shorter path) on the complex index, and the indirect index component is a variable that affects the complex index by affecting other direct indexes, and has a smaller influence on the complex index. Fish bone map or Dan Chuantu is a tool for identifying potential causes of a particular event or outcome. In determining the impact path, a causal graph may be drawn to demonstrate causal relationship links between the target index and various factors. By layer-by-layer decomposition, it is clear which are direct factors, which are indirect factors, and the interaction paths between them.
It should be noted that, the correlation degree between two nodes is calculated by using pearson correlation coefficient (Pearson correlation coefficient), which is an index for measuring the correlation degree between two variables, and the pearson correlation coefficient is calculated by using covariance and standard deviation of the variables, and the value range is [ -1, 1]. When the correlation between two variables is enhanced, the correlation coefficient tends to be 1 or-1, and when there is no relationship between the two variables, the correlation coefficient tends to be 0.
Step S103, determining interaction relation among index components according to the association relation among the index components, determining whether a coupling effect exists in the interaction relation among the index components, and analyzing and quantifying the coupling effect if the coupling effect exists.
In this embodiment, the interaction relationship and the coupling effect are not exactly the same, although they are related. The interaction relationship more broadly describes interactions between factors, and the coupling effect is more complex in particular in those parts of such interactions that result in non-linear or magnification/minification effects on the target variable. The interaction relation among the index components is determined according to the association relation among the index components, the direct index components, the semi-direct index components and the indirect index components are respectively corresponding to the respective association (the association of a plurality of index components), and data of the index components can be subjected to statistical analysis such as correlation analysis, regression analysis and the like by using a data analysis tool such as Excel, SPSS, SAS and the like so as to reveal the interaction relation among the index components.
In some embodiments of the present application, the index components are associated according to the type of the index component, and the association relationship between the index components is determined, including,
Corresponding index components are associated by direct index components, corresponding index components are associated by path routes related to semi-direct index components, and corresponding index components are associated by path routes related to indirect index components.
In this embodiment, the direct index component corresponds to a direct index, that is, a variable that directly affects a complex index, the indirect index component corresponds to a direct index and an indirect index, which have a greater influence on the complex index, and the indirect index component corresponds to a direct index and an indirect index, which have a smaller influence on the complex index.
It can be understood that, the association relationship between index components refers to which variables have association, and the corresponding relationship is the association relationship between index components.
In this embodiment, determining the interaction relationship between the index components according to the association relationship between the index components refers to determining a specific functional relationship between the variables according to the correspondence relationship between the variables, which may be implemented by regression analysis, and fitting the relationship between the index variables using linear regression, polynomial regression, logistic regression or other advanced regression models. These models can help understand how one index variable changes as other index variables change.
It will be appreciated that determining the interaction relationship may also be accomplished based on path analysis, which may be used to determine direct and indirect effects between index variables if the relationship between these variables is complex and involves multiple intermediate variables. Path analysis allows researchers to examine a pre-set causal model and estimate the intensity and saliency of each path.
In some embodiments of the application, and determining whether there is a coupling effect in the interaction relationship between the index components, includes,
Carrying out standardization processing on a plurality of variables involved in the interaction relation among index components, calculating covariance matrixes among the plurality of variables, solving the covariance matrixes to obtain characteristic values and characteristic vectors, obtaining a plurality of new variables, giving weights to the new variables through the characteristic values, and carrying out weighted summation on the new variables to obtain an initial comprehensive representative index;
Adjusting an initial comprehensive representative index through the stability of a plurality of variables involved in the interaction relation between index components to obtain a comprehensive representative index; Wherein, the method comprises the steps of, In order to synthesize the representative index of the present invention,For the purpose of initially synthesizing the representative index,To indicate the number of index members that are in an interaction relationship,To the first of the interaction relationshipStability of the individual index assembly,For a maximum stability of the index assembly in which an interaction relationship exists,The factor of the fluctuation is used to determine,For the fluctuation factor mapped from the stability maximum of the index components having an interaction relationship,Calculating the nonlinear dependency degree between the comprehensive representative index and the complex index, and judging whether the coupling effect exists or not according to the nonlinear dependency degree.
In this embodiment, in order to determine whether a coupling effect is generated or present in the interaction relationship, several variables are integrated to obtain a comprehensive representative index, and Principal Component Analysis (PCA) is used for integration.
Specifically, since the dimensions and distribution of the variables may be different, it is necessary to perform normalization processing on the data first. Since the dimensions (i.e., units of measure) and distribution (i.e., range and morphology of the data) of the variables may be different, direct analysis may lead to inaccuracy or bias in the results. Therefore, it is necessary to perform normalization processing on the data so that the variables have the same dimensions and distribution.
More specifically, after normalization, the covariance matrix of the data is calculated to reflect the correlation between the variables. Covariance matrices are a matrix that describes the correlation between variables in a dataset, reflecting the strength and direction of the system between variables.
More specifically, the principal component (new variable) is extracted by solving eigenvalues and eigenvectors of the covariance matrix. The main component is a new and mutually independent variable extracted from the original data, and can keep the information of the original data to the maximum extent. The principal direction of change of the data can be found by solving eigenvalues and eigenvectors of the covariance matrix. These directions are the principal components, and are ranked according to the magnitude of the feature values, and the larger the feature value is, the more information the corresponding principal component contains.
More specifically, the value of the comprehensive index is calculated according to the extracted principal component and the corresponding characteristic value. Based on the extracted principal components and the corresponding eigenvalues, one or more values of the overall index may be calculated. These indices are typically obtained by weighting and summing the principal components according to their eigenvalues, either by the eigenvalues themselves or by some form of transformation of the eigenvalues.
In the present embodiment of the present invention,Represents the stability class average value of a plurality of variables, adjusts the initial comprehensive representative index according to the stability class average value of the variables,The stability class average representing the plurality of variables corrects the initial integrated representative index.
In this embodiment, whether the coupling effect exists is determined by analyzing the nonlinear dependency degree of the comprehensive representative index and the complex index. If two or more variables have an interdependence in affecting a complex indicator, and if interactions between the variables result in a nonlinear relationship in the impact on the complex indicator, this is often a sign of coupling effects.
In this embodiment, the nonlinear dependency degree is a nonlinear interdependence degree, which is a method based on state space reconstruction and neighbor distance, and is used for determining the direction and the magnitude of the causal relationship. And (3) respectively representing comprehensive representative indexes and complex indexes for two independent systems or factors X and Y, and establishing state spaces of the two systems according to a state space reconstruction theory.
For sample points xn, xrn,1,.. xrn, k in state space X. Representing k adjacent points of xn in a state space X, and calculating the Euclidean distance average value of the xn and the k adjacent points;
For sample points yn, ysn,1,.. ysn, k in state space Y represent k neighboring points of yn in state space Y, map them into state space X, calculate the euclidean distance average of xn and k neighboring points xsn,1,.. xsn, k;
;
to simplify the calculation, the average distance of xn from all N sample points can be used; the nonlinear interdependence degree is a state space method, and the causal relationship between the systems judged according to the mapping relationship of the state space is defined as: and is available by definition of the term, When (when)Approaching 0, systems X and Y are independent of each other whenSignificantly greater than 0, there is a causal relationship from system X to Y, the closer to 1 the stronger the causality.
It should be noted that, the other indexes capable of characterizing the causal relationship are all the same, and the application provides only one specific mode.
In this embodiment, the coupling effect is analyzed and quantified, and may be analyzed and quantified by a preset coupling mathematical model, a regression analysis model, a causal chain model, and the like.
In some embodiments of the application, the method further comprises:
And if the coupling effect does not exist, visually displaying the index components and the interaction relation based on the mapping relation between the related business link flow and the index components.
Step S104, the index components, the interaction relation and the coupling effect are visually displayed based on the mapping relation between the related business link flow and the index components, so that attribution display of the business links is realized.
In this embodiment, the indicators, the interaction relationships, and the coupling effects are visually displayed on the relevant business link flows, which helps the decision maker to more clearly analyze the cause of complex indicator formation.
Compared with the prior art, the invention has the beneficial effects that:
1. the related business link flow and the index components are compared, the supplement of the index components is completed, and the mapping relation between the index components is constructed, so that the subsequent comparison is convenient.
2. The index components are screened and selected through redundancy to keep part of the index components, redundant index components with similar meaning are removed, the independence of the index components is guaranteed, and the reliability of analysis of the index components is improved.
3. The interaction relation among the index components is determined according to the association relation among the index components, whether the coupling effect exists in the interaction relation among the index components is determined, the association among the index components is accurately given, the interaction relation among the index components is obtained, whether the coupling effect is generated by the interaction relation is analyzed, adaptability and accuracy of complex index attribution analysis are improved, and traceability of attribution analysis is guaranteed.
From the above description of the embodiments, it will be clear to those skilled in the art that the present invention may be implemented in hardware, or may be implemented by means of software plus necessary general hardware platforms. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.), and includes several instructions for causing a computer device (may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective implementation scenario of the present invention.
Correspondingly, the application also provides a attribution analysis system applied to the complex index, as shown in figure 2, comprising,
The system comprises a first module, a second module and a third module, wherein the first module is used for acquiring a complex index model and a business target aimed at by a complex index in the financial or business field, obtaining a relevant business link flow through mapping the business target aimed at by the complex index, disassembling the complex index model according to a dimension level to obtain a plurality of index components, and comparing the relevant business link flow with the plurality of index components so as to supplement the index components and construct a mapping relation between the relevant business link flow and the index components;
The second module is used for analyzing the redundancy among the index components, screening and selecting the index components through the redundancy to reserve part of the index components, dividing the types of the index components according to the relation between the complex index and each index component, associating the index components according to the types of the index components, and determining the association relation among the index components;
A third module, configured to determine an interaction relationship between the index components according to the association relationship between the index components, determine whether a coupling effect exists in the interaction relationship between the index components, and if the coupling effect exists, analyze and quantify the coupling effect;
and the fourth module is used for visually displaying the index components, the interaction relation and the coupling effect based on the mapping relation between the related business link flow and the index components, so that attribution display of the business links is realized.
Those skilled in the art will appreciate that the drawing is merely a schematic illustration of a preferred implementation scenario and that the modules or flows in the drawing are not necessarily required to practice the invention.
Those skilled in the art will appreciate that the modules in the system in the implementation scenario may be distributed in the system in the implementation scenario according to the implementation scenario description, or that corresponding changes may be located in one or more systems different from the implementation scenario. The modules of the implementation scenario may be combined into one module, or may be further split into a plurality of sub-modules.
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art, who is within the scope of the present invention, should make equivalent substitutions or modifications according to the technical scheme of the present invention and the inventive concept thereof, and should be covered by the scope of the present invention.

Claims (7)

1.一种应用于复杂指标的归因分析方法，其特征在于，包括，1. An attribution analysis method applied to complex indicators, characterized by comprising: 获取金融或商业领域下的复杂指标模型和复杂指标针对的业务目标，通过复杂指标针对的业务目标映射来得到相关的业务环节流程，将复杂指标模型按照维度层次进行拆解，得到多个指标组件，根据相关的业务环节流程和多个指标组件进行比对，以此来补充指标组件并构建相关的业务环节流程与指标组件之间的映射关系；Obtain complex indicator models and business objectives in the financial or commercial fields, obtain relevant business process flows through mapping the complex indicators to business objectives, decompose the complex indicator models according to the dimensional levels, obtain multiple indicator components, and compare the relevant business process flows with multiple indicator components to supplement the indicator components and build a mapping relationship between the relevant business process flows and indicator components; 分析指标组件之间的冗余度，并通过冗余度对指标组件进行筛选选来保留部分指标组件，再根据复杂指标与每个指标组件之间关系来划分指标组件的类型，依据指标组件的类型来将指标组件进行关联，确定指标组件之间的关联关系；Analyze the redundancy between indicator components, and select and retain some indicator components by screening the indicator components according to the redundancy. Then, classify the types of indicator components according to the relationship between the complex indicator and each indicator component, associate the indicator components according to the type of indicator components, and determine the association relationship between the indicator components. 根据指标组件之间的关联关系来确定指标组件之间的相互作用关系，并且确定指标组件之间的相互作用关系中是否存在耦合效应，若存在耦合效应，则分析并量化耦合效应；Determine the interaction relationship between indicator components based on the association relationship between indicator components, and determine whether there is a coupling effect in the interaction relationship between indicator components. If there is a coupling effect, analyze and quantify the coupling effect; 基于相关的业务环节流程与指标组件之间的映射关系将指标组件、相互作用关系、耦合效应进行可视化展示，从而实现业务环节的归因展示；Based on the mapping relationship between the relevant business process and indicator components, the indicator components, interaction relationships, and coupling effects are visualized to achieve attribution display of business links; 其中，in, 将复杂指标模型按照维度层次进行拆解，得到多个指标组件，包括，The complex indicator model is disassembled according to the dimension level to obtain multiple indicator components, including: 根据复杂指标模型来确定复杂指标的函数，将复杂指标的函数拆解成多个变量，确认每个变量所属维度层次，在同一维度层次下，寻找与该变量相接近的若干个其它变量，计算每个变量和相接近的若干个其他变量中每个变量的稳定性；Determine the function of the complex indicator according to the complex indicator model, decompose the function of the complex indicator into multiple variables, confirm the dimensional level to which each variable belongs, find several other variables close to the variable at the same dimensional level, and calculate the stability of each variable and several other variables close to it; 通过变量的属性来分配每个变量的判定周期，整合每个判定周期下变量的变异系数来确定每个变量的稳定性，将变量的数值和其稳定性作为指标组件；Assign the judgment period of each variable by the attributes of the variable, integrate the coefficient of variation of the variable under each judgment period to determine the stability of each variable, and use the value of the variable and its stability as the indicator component; 并且确定指标组件之间的相互作用关系中是否存在耦合效应，包括，And determine whether there are coupling effects in the interaction relationships between indicator components, including, 将指标组件之间的相互作用关系中涉及的几个变量进行标准化处理，计算几个变量之间的协方差矩阵，求解协方差矩阵得到特征值和特征向量，以此得到多个新变量，通过特征值对新变量赋予权重，对新变量进行加权求和，得到初始综合代表性指标；Standardize several variables involved in the interaction relationship between indicator components, calculate the covariance matrix between several variables, solve the covariance matrix to obtain eigenvalues and eigenvectors, and thereby obtain multiple new variables. Weights are assigned to the new variables through eigenvalues, and weighted summation is performed on the new variables to obtain the initial comprehensive representative indicator. 通过指标组件之间的相互作用关系中涉及的几个变量的稳定性来调整初始综合代表性指标，得到综合代表性指标；The initial comprehensive representative index is adjusted by the stability of several variables involved in the interaction relationship between the indicator components to obtain a comprehensive representative index; 其中，为综合代表性指标，L为初始综合代表性指标，n为存在相互作用关系的指标组件的数量，Wi为存在相互作用关系的第i个指标组件的稳定性，max(Wi)为存在相互作用关系的指标组件的稳定性最大值，τ为波动因子，max(Wi)→τ为由存在相互作用关系的指标组件的稳定性最大值映射得到的波动因子，k为预设常数；in, is the comprehensive representative index, L is the initial comprehensive representative index, n is the number of indicator components with interaction relationship, Wi is the stability of the ith indicator component with interaction relationship, max( Wi ) is the maximum stability value of the indicator components with interaction relationship, τ is the volatility factor, max( Wi )→τ is the volatility factor mapped from the maximum stability value of the indicator components with interaction relationship, and k is a preset constant; 计算综合代表性指标与复杂指标之间的非线性依赖程度，通过非线性依赖程度判断是否存在耦合效应。The nonlinear dependence degree between comprehensive representative indicators and complex indicators is calculated, and whether there is a coupling effect is determined by the nonlinear dependence degree. 2.根据权利要求1所述的应用于复杂指标的归因分析方法，其特征在于，分析指标组件之间的冗余度，包括，2. The attribution analysis method for complex indicators according to claim 1 is characterized in that the redundancy between the indicator components is analyzed, including: 根据指标组件变量的数值构建其随时间变化的变量曲线图，确定每个指标组件变量的文本关键词，比对指标组件之间变量的文本关键词的相似度，从而确定具有相似文本关系的待选指标组件；According to the values of the indicator component variables, a variable curve graph of the indicator component variables changing over time is constructed, the text keywords of each indicator component variable are determined, and the similarity of the text keywords of the variables between the indicator components is compared, so as to determine the candidate indicator components with similar text relationships; 将具有相似文本关系的待选指标组件之间的变量曲线图进行比对，得到变量曲线相似度，根据具有相似文本关系的待选指标组件的变量曲线相似度、文本关键词的相似度和变量的稳定性来计算具有相似文本关系的待选指标组件之间的冗余度；Compare the variable curve graphs between the candidate indicator components with similar textual relationships to obtain the variable curve similarity, and calculate the redundancy between the candidate indicator components with similar textual relationships according to the variable curve similarity of the candidate indicator components with similar textual relationships, the similarity of text keywords and the stability of variables; P＝α1Q1+α2Q2+α3β|Q3-Q4|P＝α 1 Q 1 +α 2 Q 2 +α 3 β|Q 3 -Q 4 | 其中，P为具有相似文本关系的待选指标组件之间的冗余度，α1为具有相似文本关系的待选指标组件的变量曲线相似度的冗余权重，Q1为具有相似文本关系的待选指标组件的变量曲线相似度，α2为具有相似文本关系的待选指标组件的文本关键词的相似度的冗余权重，Q2为具有相似文本关系的待选指标组件的文本关键词的相似度，α3为具有相似文本关系的待选指标组件的变量的稳定性的冗余权重，β为转换系数，Q3、Q4为具有相似文本关系的待选指标组件的变量的稳定性。Among them, P is the redundancy between the candidate indicator components with similar textual relations, α1 is the redundant weight of the variable curve similarity of the candidate indicator components with similar textual relations, Q1 is the variable curve similarity of the candidate indicator components with similar textual relations, α2 is the redundant weight of the similarity of the text keywords of the candidate indicator components with similar textual relations, Q2 is the similarity of the text keywords of the candidate indicator components with similar textual relations, α3 is the redundant weight of the stability of the variables of the candidate indicator components with similar textual relations, β is the conversion coefficient, Q3 and Q4 are the stability of the variables of the candidate indicator components with similar textual relations. 3.根据权利要求2所述的应用于复杂指标的归因分析方法，其特征在于，并通过冗余度对指标组件进行筛选选来保留部分指标组件，包括，3. The attribution analysis method for complex indicators according to claim 2 is characterized in that the indicator components are screened and selected by redundancy to retain some indicator components, including: 若冗余度超过冗余度阈值，则将该具有相似文本关系的待选指标组件之间中变量的稳定性较低一方的待选指标组件剔除掉，保留变量的稳定性较高一方的待选指标组件；If the redundancy exceeds the redundancy threshold, the candidate indicator component with the lower variable stability among the candidate indicator components with similar textual relations is eliminated, and the candidate indicator component with the higher variable stability is retained; 否则，保留该具有相似文本关系的待选指标组件中的所有待选指标组件。Otherwise, all candidate indicator components in the candidate indicator components having similar textual relations are retained. 4.根据权利要求1所述的应用于复杂指标的归因分析方法，其特征在于，再根据复杂指标与每个指标组件之间关系来划分指标组件的类型，包括，4. The attribution analysis method for complex indicators according to claim 1 is characterized in that the types of indicator components are divided according to the relationship between the complex indicator and each indicator component, including: 指标组件的类型包括直接指标组件、半直接指标组件和间接指标组件；The types of indicator components include direct indicator components, semi-direct indicator components, and indirect indicator components; 根据相关的业务环节流程与指标组件来构建复杂指标的鱼骨图，从而确定鱼骨图中的鱼头和鱼骨，鱼头为复杂指标，鱼骨为指标组件，将鱼骨图转换成指标组件影响路径图，指标组件为影响路径图中的节点，复杂指标为路径终点；Construct a fishbone diagram of complex indicators based on the relevant business process and indicator components, thereby determining the fish head and fish bones in the fishbone diagram. The fish head is the complex indicator, and the fish bones are the indicator components. Convert the fishbone diagram into an indicator component impact path diagram. The indicator components are the nodes in the impact path diagram, and the complex indicators are the end points of the path. 计算具有连通关系的相邻两个节点之间的相关度，并通过相关度来定义两个节点之间的路径长度，将与路径终点直接连通的节点的指标组件划分为直接指标组件，并且计算节点到路径终点的路径长度，将路径长度在预设长度范围内的节点的指标组件划分为半直接指标组件，并且确定该节点到路径终点的路线；Calculate the correlation between two adjacent nodes with a connected relationship, and define the path length between the two nodes by the correlation, divide the index component of the node directly connected to the path end point into a direct index component, and calculate the path length from the node to the path end point, divide the index component of the node whose path length is within a preset length range into a semi-direct index component, and determine the route from the node to the path end point; 将其它节点的指标组件划分为间接指标组件，并且确定该节点到路径终点的路线。The indicator components of other nodes are divided into indirect indicator components, and the route from the node to the end point of the path is determined. 5.根据权利要求4所述的应用于复杂指标的归因分析方法，其特征在于，依据指标组件的类型来将指标组件进行关联，确定指标组件之间的关联关系，包括，5. The attribution analysis method for complex indicators according to claim 4 is characterized in that the indicator components are associated according to the types of the indicator components to determine the association relationship between the indicator components, including: 通过直接指标组件来将对应的指标组件关联，通过半直接指标组件所涉及的路径路线来关联对应的指标组件，通过间接指标组件所涉及的路径路线来关联对应指标组件。The corresponding indicator components are associated through direct indicator components, the corresponding indicator components are associated through the path routes involved in semi-direct indicator components, and the corresponding indicator components are associated through the path routes involved in indirect indicator components. 6.根据权利要求1所述的应用于复杂指标的归因分析方法，其特征在于，所述方法还包括，6. The attribution analysis method for complex indicators according to claim 1, characterized in that the method further comprises: 若不存在耦合效应，则基于相关的业务环节流程与指标组件之间的映射关系将指标组件和相互作用关系进行可视化展示。If there is no coupling effect, the indicator components and their interaction relationships are visualized based on the mapping relationship between the relevant business process and the indicator components. 7.一种应用于复杂指标的归因分析系统，用于实现如上述权利要求1-6任一项所述的应用于复杂指标的归因分析方法，其特征在于，包括，7. An attribution analysis system applied to complex indicators, used to implement the attribution analysis method applied to complex indicators as described in any one of claims 1 to 6, characterized in that it includes: 第一模块，用于获取金融或商业领域下的复杂指标模型和复杂指标针对的业务目标，通过复杂指标针对的业务目标映射来得到相关的业务环节流程，将复杂指标模型按照维度层次进行拆解，得到多个指标组件，根据相关的业务环节流程和多个指标组件进行比对，以此来补充指标组件并构建相关的业务环节流程与指标组件之间的映射关系；The first module is used to obtain complex indicator models and business objectives in the financial or commercial fields, obtain relevant business process links through mapping the business objectives of complex indicators, decompose the complex indicator model according to the dimensional level, obtain multiple indicator components, and compare the relevant business process links with multiple indicator components to supplement the indicator components and build a mapping relationship between the relevant business process links and indicator components; 第二模块，用于分析指标组件之间的冗余度，并通过冗余度对指标组件进行筛选选来保留部分指标组件，再根据复杂指标与每个指标组件之间关系来划分指标组件的类型，依据指标组件的类型来将指标组件进行关联，确定指标组件之间的关联关系；The second module is used to analyze the redundancy between indicator components, and select and retain some indicator components by screening the indicator components according to the redundancy, and then divide the types of indicator components according to the relationship between the complex indicator and each indicator component, and associate the indicator components according to the type of indicator components to determine the association relationship between the indicator components; 第三模块，用于根据指标组件之间的关联关系来确定指标组件之间的相互作用关系，并且确定指标组件之间的相互作用关系中是否存在耦合效应，若存在耦合效应，则分析并量化耦合效应；The third module is used to determine the interaction relationship between the indicator components according to the association relationship between the indicator components, and determine whether there is a coupling effect in the interaction relationship between the indicator components. If there is a coupling effect, the coupling effect is analyzed and quantified; 第四模块，用于基于相关的业务环节流程与指标组件之间的映射关系将指标组件、相互作用关系、耦合效应进行可视化展示，从而实现业务环节的归因展示。The fourth module is used to visualize the indicator components, interaction relationships, and coupling effects based on the mapping relationship between the relevant business link processes and the indicator components, thereby realizing the attribution display of the business links.
CN202411346425.8A 2024-09-26 2024-09-26 Attribution analysis method and attribution analysis system applied to complex indexes Active CN118885912B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202411346425.8A CN118885912B (en) 2024-09-26 2024-09-26 Attribution analysis method and attribution analysis system applied to complex indexes

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202411346425.8A CN118885912B (en) 2024-09-26 2024-09-26 Attribution analysis method and attribution analysis system applied to complex indexes

Publications (2)

Publication Number Publication Date
CN118885912A CN118885912A (en) 2024-11-01
CN118885912B true CN118885912B (en) 2024-12-06

Family

ID=93229701

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202411346425.8A Active CN118885912B (en) 2024-09-26 2024-09-26 Attribution analysis method and attribution analysis system applied to complex indexes

Country Status (1)

Country Link
CN (1) CN118885912B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119379425B (en) * 2024-12-31 2025-07-22 杭银消费金融股份有限公司 Attribution analysis method and system for overdue index change

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107818238A (en) * 2017-09-28 2018-03-20 河海大学 A kind of method for determining coupling between evapotranspiration change main cause and differentiation factor
CN111401788A (en) * 2020-04-10 2020-07-10 支付宝(杭州)信息技术有限公司 Attribution method and device of service timing sequence index

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101814958B1 (en) * 2017-06-07 2018-01-30 상명대학교산학협력단 Apparatus of analyzing causal relationship between life logging data and method thereof
US20220222594A1 (en) * 2021-01-12 2022-07-14 Adobe Inc. Facilitating analysis of attribution models
CN115994696A (en) * 2021-10-15 2023-04-21 腾讯科技（深圳）有限公司 Attribution index determining method, attribution index determining device, attribution index determining equipment and storage medium
CN114943424B (en) * 2022-04-29 2025-02-07 深圳供电局有限公司 A method and system for generating enterprise management index relationship
CN115545881A (en) * 2022-09-02 2022-12-30 睿智合创(北京)科技有限公司 Credit risk processing-based risk factor attribution method
CN115757664B (en) * 2023-01-10 2023-04-25 深圳市规划和自然资源数据管理中心(深圳市空间地理信息中心) A Causal Relationship Mining Method Between SDG Indicators by Coupling Transfer Entropy and HITS Algorithm

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107818238A (en) * 2017-09-28 2018-03-20 河海大学 A kind of method for determining coupling between evapotranspiration change main cause and differentiation factor
CN111401788A (en) * 2020-04-10 2020-07-10 支付宝(杭州)信息技术有限公司 Attribution method and device of service timing sequence index

Also Published As

Publication number Publication date
CN118885912A (en) 2024-11-01

Similar Documents

Publication Publication Date Title
CN109977151B (en) Data analysis method and system
US8577791B2 (en) System and computer program for modeling and pricing loan products
US7941363B2 (en) System, method and computer program product for managing credit decisions
US8195473B2 (en) Method and system for optimized real estate appraisal
KR100771718B1 (en) Computer-implemented inference methods, portfolio rating systems, and computers to infer the evaluation of credit scores
US20160239919A1 (en) Predictive model development system applied to organization management
US20040015376A1 (en) Method and system to value projects taking into account political risks
US20080071702A1 (en) Investment classification and tracking system
KR100746107B1 (en) Technical properties of asset portfolios Correlation methods, correlation systems, and correlation computers
Nguyen et al. Constrained fuzzy hierarchical analysis for portfolio selection under higher moments
WO2005043331A2 (en) Method and apparatus for creating and evaluating strategies
MXPA01008622A (en) Rapid valuation of portfolios of assets such as financial instruments.
MXPA01008620A (en) Valuation prediction models in situations with missing inputs.
Dia et al. Assessing the performance of Canadian credit unions using a three-stage network bootstrap DEA
CN101097620A (en) Method and system for business transformation management
MXPA01008623A (en) Methods and systems for efficiently sampling portfolios for optimal underwriting.
MXPA01008619A (en) Methods and systems for optimizing return and present value.
US20240346531A1 (en) Systems and methods for business analytics model scoring and selection
CN118885912B (en) Attribution analysis method and attribution analysis system applied to complex indexes
US20090164297A1 (en) Integrated business decision-making system and method
Pakšiová et al. Capital maintenance evolution using outputs from accounting system
Shi et al. A big data analytics method for assessing creditworthiness of SMEs: Fuzzy equifinality relationships analysis
Kumra et al. Assessing a knowledge-based approach to commercial loan underwriting
Islam Predictive capability of financial ratios for forecasting of corporate bankruptcy
Mehlawat Behavioral optimization models for multicriteria portfolio selection

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant