Financial Risk Prediction For Listed Companies Based On PCA-BP And SVM

With the globalization of the economy and the intensification of market competition,listed companies are facing increasingly severe challenges.The control of financial risks has been increasingly valued by listed companies and has become a hot issue in the economic field.In order to effectively prevent the risks of listed companies,financial risks are predicted.Systematic and non-systematic risk factors are considered in the analysis process.Approaches like principal component analysis,BP neural network and SVM are incorporated.The contributions of this thesis are as follows:(1)A financial risk index system for listed companies is constructed by considering systemic financial risk factors.The BP neural network is used to model and forecast risks.,Compared to the analysis under non-systematic risks,the accuracy of financial risk forecasting is significantly increased when considers systematic risks.(2)A financial risk prediction model based on PCA-BP neural network is constructed.The PCA method is used to simplify the BP network structure,using Matlab software programming.Empirical analysis is performed on 276 financial data samples of listed companies from 2013 to 2014.Results show that the accuracy of risk prediction based on PCA-BP neural network model is significantly higher than that of the traditional BP neural network model.(3)A financial risk prediction model based on SVM is constructed R software is used to perform the empirical analysis of 38 financial data from 2014.Results show that the SVM model is more accurate than the PCA-BP model method.The rate has been improved.In addition,the SVM model supports small sample predictions.In this thesis,financial risk index system,PCA-BP model and SVM model are developed.The actual data from the Guotai'an database is used to predict the financial risks of listed companies.This study provides a quantitative basis for the listed companies to identify and avoid financial risks.In addition,it will help listed companies develop in a healthy environment.