Construction of Comprehensive Evaluation Model of Credit Rating of Small and Medium Enterprises and Empirical Research

From the perspective of analysis of big data, based on RESSET database of financial research, authors of this article choose quarterly financial index data of 60 companies in 20 quarters on small and medium board, combining with the selected non financial indicators and use principal component analysis, entropy method, panel exponential smoothing and the analytic hierarchy process (AHP) to construct the comprehensive evaluation model. The survey and empirical evidence have demonstrated that credit rating system designed in this paper can preferably reflect and predict the enterprise's credit rating, and the distribution of credit rating of sample enterprises basically agrees with actual situation in morphology and grade center.