Credit Rating Model Based On Default Identification:Empirical Evidence From Chinese Small Construction Enterprises

Cash flow decides survival lifeline of small construction enterprises. In the situation of credit risk happened frequently and bank credit tightening, banks are more cautious of loaning for small construction enterprises. However, as the size of the construction enterprises is small, it is difficult to find the classic indicators and credit rating theory to evaluate the credit status of small construction enterprises, for these reasons, small construction enterprises face serious problem of the finance and loan, it is a key issue for small construction enterprises to solve credit evaluation.The small construction enterprise credit rating based on the default status consists of three parts:firstly, the establishment of the indicator system of the small construction enterprise credit rating; secondly, the small construction enterprise rating model; thirdly, the grade division of the small construction enterprise credit. The establishment of the indicator system of the small construction enterprise credit rating refers to establish an indicator system which not only reflects the liquidity of the small construction enterprise, but is capable of markedly distinguishing the default of the small construction enterprise. The small construction enterprise rating model is the one which markedly influences the default status. The grade division of the small construction enterprise measures the credit risk on the basis of the proportion of the receivable but uncollected principal interest to the receivable principal interest in order to reflect the clients’loan losses. To divide the credit grade according to the principle that the lower of the credit grade, the higher the respective default loss ratio.The paper consists of five chapters. Chapter One is the introduction; Chapter Two is the establishment of the indicator system of the small construction enterprise credit rating; Chapter Three is the small construction enterprise credit rating model on the basis of the default status; Chapter Four is the grade division of the small construction enterprise credit based on the small sample expansion; Chapter Five is the conclusion and suggestion.The contributions of this paper are in three aspects.(1) The contributions of combined weighting:It maximizes the value for the goal programming function, to ensure we maximize the difference in credit scores between good and bad customers and are able to clearly differentiate between good and bad customers.It creates the non-linear goal programming function by minimizing the sum of squares between types s,, and maximizing the sum of squares within each type sa. The smaller s,, the smaller the difference between credit scores for customers of the same type. The larger s0, the larger the difference in average credit scores between types. The smaller s,, and the larger s0, the larger the value for the goal programming function. As such, we maximize the value for the goal programming function, to ensure we maximize the difference in credit scores between good and bad customers and are able to clearly differentiate between good and bad customers.(2) The contributions of small sample expansion:generate new data by Bootstrap interpolation method, use copula function to simulate the correlation between new data, thus expand the data sample of three indicators at the same time, the credit score, principal and interest receivable and the unpaid principal and interest yearly.Through Bootstrap interpolation method to generate new data, using copula function to simulate the correlation between new data, this paper expand the data sample of three indicators at the same time, the credit score, principal and interest receivable and the unpaid principal and interest yearly. Thus we generate large sample that is enough to undertake credit rating, and guarantee the correlation relationship between new data of these three indicators as well, solving the problem that small amount of building enterprise can’t do credit rating, making up the existing research methods of multiple index data samples expansion.(3) The contributions of indicator screening:Greater mean deviation of default sample from the whole sample lead to bigger deviation from non-default sample as well, and the indicator can easily distinguish default and non-default sample.The indicator i data is divided into two categories, default and non-default. Greater mean deviation of default sample from the whole sample lead to bigger deviation from non-default sample as well, and the indicator can easily distinguish default and non-default sample. In our study we select the indicator for credit rating indicator system which identify whether it is default or non default using F test. This avoids the disadvantages of the existing research in which the standard has nothing to do with the default selection indicator, and paves a new way of credit rating.