Automatic Spline Knot Selection in Modeling Mortgage Loan Default Using Shape Constrained Regression

2021 ◽  
pp. jsf.2021.1.123
Author(s):  
Guangning Xu ◽  
Geng Deng ◽  
Xindong Wang ◽  
Ken Fu
2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Azira Abdul Adzis ◽  
Hock Eam Lim ◽  
Siew Goh Yeok ◽  
Asish Saha

PurposeThis study investigates factors contributing to residential mortgage loans default by utilizing a unique dataset of borrowers' default data from one of the pioneer lending institutions in Malaysia that provides home financing to the public. Studies on mortgage loan default have been extensively examined, but limited studies utilize the individual borrower's data, as financial institutions generally hesitant to reveal their customers' data due to confidentiality issue.Design/methodology/approachThis study uses logistic regression model to analyze 47,158 housing loan borrowers' data for the year 2016.FindingsThe findings suggest that male borrowers, Malay and other type of ethnicity, guarantor availability, loan original balance, loan tenure, loan interest rate and loan-to-value (LTV) ratio are the significant factors that influence mortgage loans default in Malaysia.Research limitations/implicationsFuture studies may expand the sample by employing data from other types of financial institutions that would give greater insights as findings might vary due to differences in objectives, functions and regulations. In addition, the findings are subjected to the censoring bias where future studies could perform the survival analysis to control for censoring bias and re-validating the findings of the present study.Practical implicationsThe findings provide valuable insights for lending institutions and the government to formulate housing loan policy in Malaysia.Originality/valueTo the best of the authors' knowledge, this is the first study in the context of emerging economies that uses financial institution's internal data to investigate factors of mortgage loan default.


Author(s):  
De-Graft Owusu-Manu ◽  
Richard Ohene Asiedu ◽  
David John Edwards ◽  
Kenneth Donkor-Hyiaman ◽  
Pius Akanbang Abuntori ◽  
...  
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document