enterprise risk trust loss model based on Cox regression prediction

Abstract

The invention discloses an enterprise risk trust loss model based on Cox regression prediction. the lost credit survival probability of the lost credit model is calculated; f (D), using an enterprisecredit model feature Y as a covariable or an interaction item; The construction method of the lost credit model comprises the following steps of: constructing a lost credit model; Q1, determining a feature Y of the lost credit model, Q2: formulating a lost credit model observation starting time D1; Q3: formulating a lost credit observation time D3; Q4: formulating a lost credit end point time D2 or D3; Q5, formulating a lost credit survival time D; Q6, determining a reference risk function f0 (D) of the lost trust model; Q7, determining a partial regression coefficient of the characteristic Yof the lost trust model through likelihood estimation; Q8, brining The survival time D, the untrustworthy model feature Y, the untrustworthy model reference risk function f0(D), and the partial regression coefficient Beta of the untrustworthy model feature Y into the unbalanced risk function formula of the Cox proportional regression model. According to the model disclosed by the invention, the future risk change trend of an enterprise can be predicted, so that the possibility of enterprise credit risk occurrence can be predicted. The model disclosed by the invention has the advantages that the risk change trend of the enterprise can be predicted.

Description

technical field [0001] The invention relates to the field of information security, in particular to a Cox regression prediction-based enterprise risk and dishonesty model. Background technique [0002] In the fierce market competition, enterprises are always facing risks brought by changes and uncertainties in the external market environment and internal management operations. If there is no timely warning and control, the risks will turn into crises, which will lead to the failure of business operations. even bankruptcy. In recent years, enterprise risk assessment has been paid more and more attention by enterprises, governments, and investors, and risk assessment has been added to enterprise management and investment decision-making as a key focus factor. Since corporate risks are affected by dynamic indicators such as corporate strategy, finance, operations, and macro-environment, the quantitative assessment results of their dishonesty risks often change dynamically over...

AI Technical Summary

Problems solved by technology

Enterprises, especially investors, pay attention to the future trend of enterprise risk changes in order to foresee the possibility of enterprise risk occurrence, while the existing credit risk assessment is to evaluate the current enterprise risk based on historical data, which is often difficult to meet the needs of enterprises, especially investors

Images