| Summary: | 碩士 === 銘傳大學 === 財務金融學系碩士在職專班 === 102 === Lending is still the main profit source for banks now. And banks’ making home loans brings in a stable income with the lowest default risk. The present loose monetary policy makes the money in the market abundant and the interest rate is low, which makes the loan-deposit interest margin of the banking industry unable to be expanded. Therefore, how to increase the banks’ chance of making profits while the income from the home loans is stable has become a topic for the banking industry to ponder on. And the home equity loan emerges just in such an environment.
This research studies 61,182 households that applied for home equity loans successfully from 2008 to 2011, where 61,080 households paid on time and 102 households paid late. It conducts an empirical analysis on 10 risk variables of households with home equity loans using the logistic regression model to find out important risk variables and discuss the relationship between important risk variables and whether the home equity loan is overdue. It expects to really reflect the loan quality, hoping to be used as an important reference for the banking industry to make credit policy and loans. The research results show that educational background, gender, age, marital status, type of households with loans and real estate location are important risk variables. If house owners are male, it is significantly positively related to whether the home equity loan is overdue; if households with loans have applied loans before, it is significantly negatively related to whether the home equity loan is overdue; age is significantly positively related to whether the home equity loan is overdue; for married house owners with children, it is significantly negatively related to whether the home equity loan is overdue; if the educational background of owners is general and vocational high school and below, it is significantly positively related to whether the home equity loan is overdue; if the real estates are located in northern region, central region and southern region, it is significantly negatively related to whether the home equity loan is overdue.
After conducting a discrimination analysis on the empirical results, it finds that the important risk variables used to distinguish households that paid on time from those who paid late are gender, type of households with loans, educational background, real estate purpose and real estate provider; and gender, type of households with loans and educational background are the same with those in the empirical analysis of the logistic regression model.
|
|---|
