Font Size: a A A

A Study Of SVM-based Financial Risk Early-warning Model For China Listed SMEs

Posted on:2015-02-21Degree:MasterType:Thesis
Country:ChinaCandidate:X Y ZhuFull Text:PDF
GTID:2269330425463076Subject:Business management
Abstract/Summary:PDF Full Text Request
Small and medium-sized enterprises as an important part of our national economy,plays a very important role in China’s economic development, and also have anenormous impact on the macroeconomic growth. Compared with large enterprises,small and medium-sized enterprises facing to rapidly changing external environmentand the difficult financing environment, their financial risk consciousness andfinancing ability are relatively weak, so establishing a set of scientific financial riskanalysis and rating system to find the potential risk in time and take measures toavoid or reduce the unnecessary loss is very important. The purpose of this study, ishoping to come out a set of suitable system for analyzing and rating financial risk ofChina’s small and medium-sized manufacturing listed company, then providingscientific decision for the enterprise and its stakeholders.This paper based on the relevant literature and a review of financial risk analysistheory, selected scientific and practical financial risk rating as rating index, and theindex based on two indicators, two indicators finally got significant pretreatmentindicators which used in the model. Depending on the selected financial risk ratingindex, this paper constructed the the small and medium enterprise financial riskratings listed SVM model, and used the traditional Logit regression models ascomparative study.Studies show that the financial risk for SMEs constructed and rating model hassome practical value, although this paper used SVM model and its accuracy was84%, compared to a similar study, the accuracy rate was not high, as the contrastLogit regression model accuracy rate of83%for two years comprehensive, butconsidering the empirical results and false positives rate, support vector machinemodel accuracy rate should be slightly higher than the Logit regression model,which was consistent with other researches. This indicates that the SVM model ismore applicable to financial risk rating.
Keywords/Search Tags:financial risk and rating, support vector machine (SVM), manufacturing small and medium-sized enterprises (smes), listed SMEs
PDF Full Text Request
Related items