The Research On Credit Rating Of Science And Technology Small And Medium-Sized Enterprises

By 2014,the number of SMEs in China has reached 343135,accounting for more than 98% of all enterprises and contributing more than half of GDP.They play an important role in the entire national economy.Under the new normal,China’s economic growth tends to be gentle.In order to achieve the fast and steady economic growth in China,China’s economy needs to transform,update and change the mode of economic development.Science and technology is the first productivity,the development of China’s science and technology SMEs has injected new vitality into the development of China’s economy.The development of science and technology SMEs is essential to the development and growth of China’s economy.However,financing is always a difficult problem faced by science and technology SMEs.According to the result of credit rating,banks can determine whether to release loans to the enterprises.For this reason,on the basis of previous studies,this paper focuses on the two credit rating methods of combination evaluation and Lasso-Logistic econometric model on science and technology SMEs.First of all,we analyze the importance of science and technology small and medium-sized enterprises to the national economic development and the important issues that hinder their development.And then analyze the significance of credit rating from two angles of theory and reality.In the next place,we compare and summarize the domestic and foreign research literatures about the credit rating of small and medium-sized enterprises and elaborate the basis theory of credit rating,providing theoretical basis for the follow-up credit rating research.Secondly,based on the existing literatures,we establish a set of suitable credit rating index system for small and medium-sized enterprises combining with their characteristics.The rating system includes 11 first grade indicators and 36 secondary level indicators.On this basis,using entropy method,DEA method,equal-weight method and principal component analysis method to rate and sort for 86 samples.The four methods’ ranking results are different.In order to get more accurate,robust and consistent ranking results,this paper uses four different combination evaluation methods respectively to do credit rating.The four methods include the average value method,Borda method,Copeland method and fuzzy Borda method.After the ante test and the ex post test,we select the optimal combination method for credit rating of science and technology small and medium-sized enterprises.In addition,considering that dependent variable is the binary-valued variable and independent variables are numerous,this paper introduces Lasso algorithm to Logistic regression and constructs the Lasso-Logistic econometric model to study the credit rating of small and medium-sized enterprises.On the one hand,Lasso algorithm can deal with theissues that have tons of data,which is fit for the trend of big data.On the other hand,through the Lasso-Logistic econometric method,we can select the key variables of credit rating of small and medium-sized enterprises and estimate the parameters simultaneously.The result shows that the model’s prediction accuracy is higher than the other method.If this method were taken into practice,the manpower,material resources and the time cost of the credit rating can be sharply reduced.Furthermore,it can help investors select companies which have good credit and are suitable for long-term investment.Finally,we compare the two methods of combined evaluation method and Lasso-Logistic econometric method from two aspects of degree of difficulty and accuracy.And we analyze applicability of the two methods.The results show that the combination evaluation method is simpler and the method’s operation is easier than Lasso-Logistic econometric method.According to the results of this paper,we put forward the corresponding policy recommendations for science and technology SMEs to improve their credit rate.Last but not least,we put forward some prospects on credit rating,aiming at enriching the research about credit rating.