Online P2P lending is a new mode of financial services supported by Internet technology and it has a bright prospect. Since Online P2P lending emerged from UK in 2005, it has spread rapidly across the globe and has been concerned by numerous scholars after years of development. In China, Online P2P first appeared in 2007 and it shows a strong vitality and rapid development in just eight years. Especially, hundreds of lending platform nationwide has been built during the past two years.Online P2P lending not only helps to solve the traditional financing difficulty for small enterprises, but also meets the demand of microcredit financing for consumers. However, there is a problem of loan default rates on the high side in current P2P loan. Therefore, recognizing those factors which affect credit risks is of great significance for online lending and risk management. Based on the predecessors in the field of exploration, this paper puts forward a P2P network research model of credit risk and has carried a empirical research on P2P loans.Beginning at proposing a model for P2P networks borrowing program, the work moves to write data and then obtain the actual transaction data from P2P website. Next, using Logistic regression model, the study carried an empirical research on key factors affecting the borrower’s credit risk and its influence. The result shows that factors such as rate、sex would have influence on credit status. Finally, according to result of empirical research, several reasonable proposals have been put forward to improve both the Online P2P credit system and the operational efficiency of Online P2P platform.The study has statistical significance in solving the current P2P lending problems, which could provide some suggestions for the development of online P2P lending. |