An Application of Robust Method in Multiple Linear Regression Model toward Credit Card Debt

Credit card is a convenient alternative replaced cash or cheque, and it is essential component for electronic and internet commerce. In this study, the researchers attempt to determine the relationship and significance variables between credit card debt and demographic variables such as age, househo...

Full description

Saved in:
Bibliographic Details
Published in:Journal of physics. Conference series 2018-04, Vol.995 (1), p.12011
Main Authors: Azmi, Nur Amira, Rusiman, Mohd Saifullah, Khalid, Kamil, Roslan, Rozaini, Sufahani, Suliadi, Mohamad, Mahathir, Salleh, Rohayu Mohd, Amir Hamzah, Nur Shamsidah
Format: Article
Language:English
Subjects:
Credit cards
Demographic variables
Education
Electronic commerce
Family income
Income
Physics
Regression analysis
Regression models
Robustness (mathematics)
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Credit card is a convenient alternative replaced cash or cheque, and it is essential component for electronic and internet commerce. In this study, the researchers attempt to determine the relationship and significance variables between credit card debt and demographic variables such as age, household income, education level, years with current employer, years at current address, debt to income ratio and other debt. The provided data covers 850 customers information. There are three methods that applied to the credit card debt data which are multiple linear regression (MLR) models, MLR models with least quartile difference (LQD) method and MLR models with mean absolute deviation method. After comparing among three methods, it is found that MLR model with LQD method became the best model with the lowest value of mean square error (MSE). According to the final model, it shows that the years with current employer, years at current address, household income in thousands and debt to income ratio are positively associated with the amount of credit debt. Meanwhile variables for age, level of education and other debt are negatively associated with amount of credit debt. This study may serve as a reference for the bank company by using robust methods, so that they could better understand their options and choice that is best aligned with their goals for inference regarding to the credit card debt.
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/995/1/012011