Modeling on Human Development Index in East Java Using M-Estimation Robust Regression
Qodharu Aulia Wahyu Purnama, Yuliana Susanti, and Hasih Pratiwi

Department of Statistics, Faculty of Math and Science, Universitas Sebelas Maret, Indonesia
a) qodharu[at]gmail.com
b)yulianasusanti[at]staff.uns.ac.id
c)hpratiwi[at]mipa.uns.ac.id

Abstract

Human Development Index (HDI) is one of the indicators for estimating the effectiveness in improving the quality of human life. Therefore, HDI can determine the level of development in an area. Three basic dimensions that create HDI are health, education, and living standards used as the human life quality standards. East Java is one of the provinces in Indonesia which has a good population development. This research analyzes what factors significantly influence HDI in East Java in 2020. These data have outliers, so the least-squares method is inappropriate for estimating the regression parameters. Robust regression is a method for estimating parameters when there are outliers, or residual distribution is not normal. The result showed that the robust regression model with M-estimation has several significant variables in the Human Development Index: Expected Years of Schooling, Expenditures Per Capita, and Life Expectancy. It also obtain R-SquareAdj = 0.9904.

Keywords: Human Development Index, Regression model, least square method, M-Estimation

Topic: Mathematics Education