MODEL HUBUNGAN JUMLAH PENGANGGURAN DAN INDEKS KEDALAMAN KEMISKINAN DI PULAU SUMATERA TAHUN 2019 MENGGUNAKAN REGRESI NONPARAMETRIK SPLINES

Abstract

Poverty does not only focus on decreasing the number of poor people. There is an important thing that must also be considered, namely the Poverty Gap Index (P1). From year to year, the poverty gap index for all regencies/cities in Sumatra tends to stagnate. While the island of Sumatra is the second island with the largest population in Indonesia. This should be a serious concern for the government. One of the factors that influence the poverty gap index is unemployment. The more people who are unemployed can increase the poverty gap index. Therefore we need to model the relationship between the number of unemployment and poverty gap index. The approach used is nonparametric regression modeling where the residual value is not normally distributed. The model is smoothing splines regression and quantile splines regression (median, τ = 0, 5). Meanwhile, to see the best model performance by looking at the RMSE values of both models. From the results of the study, it was found that the quantile regression smoothing splines model was better because the RMSE value was lower than the regression smoothing splines.

Keywords: poverty gap, unemployment, quantile regression

JEL Classification: I32, J64, C21