Research On Dynamic Credit Evaluation Of Certified Constructor

The Certified Constructor credit loss is a serious problem of construction market to be solved. There is an urgent need to establish a Certified Constructor credit governance system to effectively regularize their behaviors. Traditional credit evaluation model can only aggregate static index data at the isolated time points. It cannot fully mine credit behavior history record or reflect trend of credit standards, easily lead to distortion of the evaluation result.Taking the Certified Constructor as research object, the dissertation explained the credit behavior choice with the help of economics theory and behavioral economics theory. An index system for credit behavioral scoring was built which included the past behavior records of Certified Constructor in social activities and practice activities. By using entropy method to determine the index weight and time weight, a dynamic credit behavioral scoring model for Certified Constructor was built. Introduced "incentive and punishment" into the evaluation process, the dissertation set the positive and negative control line of credit to build a comprehensive evaluation model with dynamic characteristics for Certified Constructor. Based on past credit evaluation results, the dissertation built a credit rating transition probability matrix to predict certified constructors' credit trends by using Markov transfer matrix method.The results showed that:(1) the reliance on early decisions will affect certified constructors' current behavior, while the presence of the incentive effect will have a positive impact on their future behavior;(2) compared to traditional credit scoring models, the credit behavioral scoring model based on ideal point method for Certified Constructor works well in its efficiency and discrimination, easily to identify and judge the credit risk by making full use of historical information;(3) The Certified Constructor credit dynamic evaluation model can transform discrete information into continuous information and highlight the dynamic nature; taking the valuators' preferences fully into account, providing a wealth of intermediate evaluation information, being a good tool for construction administration department to use in the management;(4) the Certified Constructor credit rating transition prediction model can well reflect the credit trends, making up for the lack of long-term effect when using static models; the Certified Constructor with poor current credit level has high risk of credit deterioration, which should be the focus of government regulation.