Personal credit risk measurement model based on PSO-BP

Abstract

The invention provides a personal credit risk measurement model based on PSO-BP, and relates to the technical field of finance. The personal credit risk measurement model based on PSO-BP comprises a PSO-BP model, the PSO-BP model is based on a neural network and a particle swarm optimization algorithm, the BP neural network algorithm comprises a signal forward transmission part and an error reverse transmission part, the number of neuron nodes of an input layer is supposed to be n, the number of neuron nodes of a hidden layer is supposed to be s, and the number of neuron nodes of an output layer is supposed to be 1, according to the principle of the PSO algorithm, Vi = (vi1, vi2,... vin) and Xi = (xi1, xi2,... xin) are used for representing the speed and the position of an ith particle in an n-dimensional space respectively, the fitness function value of each particle is evaluated in each iteration, the optimal position pbest of an individual passed by each particle at the t moment and the optimal position gbest of a whole group are judged, and the optimal position pbest of the whole group is calculated. And each particle in the group updates the own speed and position according to the two optimal positions. Effective early warning signals can be provided for commercial banks, market risks can be prevented, and the method has good application value.

Description

technical field [0001] The invention relates to the field of financial technology, in particular to a PSO-BP-based personal credit risk measurement model. Background technique [0002] Credit risk is one of the most important risks faced by financial institutions. Following the traditional proportional analysis and subjective analysis, statistical methods have been widely used, such as discriminant analysis, logit regression analysis and so on. Since the late 1980s, artificial intelligence technology has been applied to credit risk measurement. At present, the most widely used in this field is the BP neural network, and its nonlinear mapping ability makes it have unique advantages. However, the parameter setting of BP neural network is based on the local information of the parameter space, and it is easy to fall into the local minimum point, which will reduce the convergence speed and prediction accuracy. Contents of the invention [0003] (1) Solved technical problems ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the deficiencies of the prior art, the present invention provides a personal credit risk measurement model based on PSO-BP, which solves the problem of low prediction accuracy

Images