Abstract：
In order to improve the precision of PPP financing risk prediction of low rent housing, this model is put forward low rent housing PPP financing risk prediction model, which is based on the PCA-SVM. First, principal component analysis (PCA) was applied to reduce the dimension of the indexes of low-rent housing PPP financing risk to eliminate redundant between indicators; and then the reduced dimension is used as a support vector machine (SVM) input, and the using of support vector machine can help to complete the prediction of low rent housing PPP financing risk. Finally combining with engineering examples proved that PCA-SVM model can effectively improve the generalization ability, the training speed and precision of support vector machine (SVM).
