Analysis of Access to Financial Services on Poverty Alleviation with MARS Approach

Abstract

This study aimed to determine the barriers of public access to financial services and their effects on poverty alleviation. The sample used was 6 ASEAN countries (Indonesia, Singapore, Malaysia, Vietnam, Thailand, and the Philippines) from 2006 to 2015. The analytical method used was the MARS. MARS is one of the nonparametric regression methods as an alternative to the multiple linear regression method, which must fulfill parametric assumptions. The results of the study using MARS show that the model formed has a high coefficient of determination, and criteria of the test of the suitability of the model are met. In other words, multivariate adaptive regression spline (MARS) can explain well the variability of the independent variables on the dependent variable. The results of the hypothesis testing using the MARS method show that indicators of macroeconomic, social, bank characteristics, institutions, and regulations affect access to financial services (AFS) and AFS affect poverty alleviation. This finding shows that increasing AFS will affect poverty reduction, and to increase public AFS can be done by minimizing macroeconomic, regulatory, social, bank, and institutional constraints.