Research On Prepayment Behavior Of Mortgage-backed Securitization In China

People began to focus on mortgage-backed securitization (MBS) again, which is one of the most important financial innovations nowadays, when the sub-prime crisis swept the world in 2007. It is developing in high-speed, and increases the liquidity of the assets, expands investment channels. Besides, MBS can promote the development of the real estate market and the securities market. Otherwise, it also provides an opportunity to spread the risk and enlarge. Although MBS is young in our country, in this context, it is necessary for our country to study MBS in-depth in order to avoid the crisis from happening again.Firstly, this paper describes the concept, origin, the basic principle and operation mechanism, then establish the quantitative model of MBS from the perspective of issuers with Operation Research method, to help investment banks determine the maximum benefits and avoid the risk of sub-prime mortgages. In addition, review the MBS process of our country, and the main risk, including prepayment risk, default risk, rate risk, and legal risk.Secondly, this paper defines pay factor, and focus on analysis of prepayment risk by pay factor; describe the threat of prepayment behavior for originator of securities, issuers, investors, real estate market and stock market. Then, put forward the factors affecting our prepayment, including loan interest rates, the real estate market prices, per capita disposable income, economic cycles and consumer attitudes, etc.; alsoFinally, this paper introduces measure indicators of the prepayment behavior, and the foreign prepayment behavior prediction models. The article points the models are not suitable for our MBS prepayment behavior by a simple evaluation; and then put forward multiple regression model based on Proportional Hazard Model, and make an empirical research to the factors affecting our prepayment of our country, which point out that the per capita disposable income, loan interest rates, housing price have a significant impact on prepayment behavior, meanwhile, shows the multiple linear regression model can be used to forecast in our MBS prepayment behavior.