As one of the main loans of commercial banks, it will bring about a great negative effect on the stability of financial system and stable operation of macro economy if the large scale of personal house mortgage loan breach takes place. By analyzing the real data of personal house mortgage loan of China commercial banks, we sort out about 15 indices on such factors as personal basic conditions, personal credit positions and loan contract which will affect the performance of the contract. On the basis of this, we use the MCLP model to set up a risk model of personal house mortgage loan and compare it with the prediction results of traditional Logistic model, and we find that the former has a higher degree of accuracy, and finally we put forward some proposals based on this conclusion.