|
|
| |
| Modeling of Human Development Index (HDI) in Indonesia Using Multivariate Adaptive Regression Spline Method
Download PDF
|
M. Aqil Zaidan Kamil, Ardi Kurniawan, Dita Amelia
|
Abstract:
Human development is one of the important aspects for the development of a country. Indonesia has a human development index value of 71.92, which is in the upper middle category in the standard category set by the Human Development Index. Thus, Indonesia still has to improve its human development index because it is still not in the high category and is still inferior to other countries. In the sustainable development goals, the Human Development Index is included in the 3rd and 4th goals, namely a healthy and prosperous life and quality education. The purpose of this study focuses on describing and modeling the Human Development Index in Indonesia, as well as interpreting the results of the best model obtained. The method used is a method with a nonparametric regression approach, namely Multivariate Adaptive Regression Spline. The results showed that the best model obtained was in a combination of 14 basis functions, a maximum interaction of 1, and a minimum observation between knots of 1. From this model, the highest level of importance was obtained for the health complaint variable with an importance level of 100%. The best MARS model produced a Generalized Cross Validation value of; 8,736 R^2 of 0.688 and a Mean Square Error of 5,402. The research is expected to be a consideration for efforts to improve the Human Development Index and support the Sustainable Development Goals (SDGs) action plan.
|
|
|
| | |