Summary: | 碩士 === 輔仁大學 === 應用統計學研究所 === 96 === The behavior of borrowers’ pre-payment makes the cash flow and the revenue of interest instable. It also makes the due date of mortgage uncertainly. Furthermore, the uncertainty of the MBS-holders’ value rises.
What kinds of borrowers will choose the loan prepayment? When did borrowers choose pre-payment? This is the interesting insures in Banks and Home Finance Agency. Therefore, this study uses Cox’s regression model of survival analysis to analyze when borrowers chose pre-payment. This study also discusses which factors will affect the borrowers to do the pre-payment. Moreover, this study adopted two step method of Decision Tree along with Logistic Regression to refine important variables of influence prepayment.
By Cox’s regression model of survival analysis, it provides the prediction of the time of prepayment. It can help the banks solve the crisis. The ability of identification in the Cox’s regression model has 80 percent in the cut point which is 0.5 and 0.7099. Ahead of borrower will choose pre-payment of the model prediction has 78.69 percent. Otherwise, the model segment client of prepayment in the future two years from the client of without prepayment. This model tries to make some warning measures to these clients. The measures include the communication with the customers and the introduction of other products. The result of this study shows that the variable of influence prepayment includes as follows: adjustable rate style, building address, the standard of appraisal, Benchmark of interest rate, delay record, mortgage premium, (prepayment) penalty rate, installment amt, the deadline of abridge prepayment.
|