Research On Individual Social Credit Based On SEM And HMM Model

In view of the current social development trend,personal credit is playing an increasingly important role in production,work and other aspects of life,so the research value of various problems in the credit system is also increasing.As a developing country,promoting credit construction can help maintain a good social atmosphere on the one hand and promote the development of all walks of life on the other hand.The paper consider the influence factors of dishonest ACTS as well as the transfer of the credit status as the starting point of two research,the study of hidden variables as a whole the ideas of the article,the first part of a structural equation model was used to study the account number,personal assets,trading and turnover of this a few hidden variables and the correlation of dishonest ACTS the hidden variables,analyses for dishonest ACTS,borrowers and turnover situation the influence degree of the largest,and the borrower's trading negative correlation with turnover situation and faithless,the borrower's personal assets and account number on the impact of personal credibility is relatively small,Moreover,the number of personal assets and accounts of the borrower is positively correlated with the situation of breach of trust.Second part using cross-sectional data to construct A set of time series,based on the HMM model is applied to the analysis of the personal credit conversion problem,and it is A category by the HMM model for 5 credit state transition matrix,the analysis of credit state transition matrix is in A state of five credit situation best A and the worst credit state E for two more stable credit state said the credit status of this part of the people in reality are relatively stable,while the remaining three state for high credit status and low credit status between the transition state.The innovation of the article lies in: in the analysis of personal credit,on the basis of the existing indicators,to join A broader range of personal indicators,such as social data,behavioral data and other social indicators;In the process of credit analysis,the essay introduces the concept of hidden variables.For example,in the first part of the article,several variables of similar nature are integrated into a hidden variable.In the second part of the article,the personal credit status is regarded as an unobservable hidden variable.When using the HMM model to analyze the credit situation,the idea of pseudo-time series is introduced,which provides a feasible way to analyze the credit state transition under the condition of only cross section data.