APPLICATION OF SEMIPARAMETRIC SPLINE REGRESSION MODEL IN MODELLING HUMAN DEVELOPMENT INDEX IN 2016 Sumi Amilia, Nar Herrhyanto, Entit Puspita
Universitas Pendidikan Indonesia
Abstract
Human development index is one of indicators to determine developed country. According to the data from Central Bureau of Statistics in 2016, human development index of Papua, West Papua, and East Nusa Tenggara, are lower than other provinces. Therefore, the aim of this study was to analyze the relationship between human development index and four variable such as life expectancy, school life expectancy, average school duration, and average per capita expenditure affecting an improvement of human development index. Data of human development index and 4 variables used in this study was obtained from the article Human Development Index 2016 published by Central Bureau of Statistics. Method used in modelling human development index in 2016 was semiparametric spline regression method. Linear spline, quadratic spline, and cubic spline were models applied in this study, whereas one point of knot was implemented as well. As a result, all variables significantly affected the human development index in Papua, while in West Papua, variables of both average school duration and average per capita expenditure related to human development index significantly. In contrast to Papua and West Papua, three variables such as school life expectancy, average school duration, and average per capita expenditure contributed on human development index in East Nusa Tenggara.
Keywords: Human Development Index, Semiparametric Spline Regression Model, Linear Spline, Quadratic Spline, Cubic Spline
