PENERAPAN METODE GEOGRAPHICALLY WEIGHTED PANEL REGRESSION (GWPR) PADA KASUS KEMISKINAN DI INDONESIA

Shantika Martha,Yundari Yundari,Setyo Wira Rizki,Ray Tamtama
DOI: https://doi.org/10.30598/barekengvol15iss2pp241-248
2021-06-01
Abstract:To analyze the factor affecting poverty during several periods by considering some geographical factors, we can use a geographically weighted panel regression (GWPR) method. GWPR is a combination of the geographically weighted regression (GWR) model and the panel regression model. The research conducts to identify the factors affecting the percentage of poor people in 34 provinces in Indonesia during 2015-2019. The results show that a suitable GWPR model is a fixed-effect model (FEM) with an exponential adaptive kernel function. Referring to the model, the province is divided into four groups based on variables having a significant effect on the percentage of poor people. That factors causing the poor people percentage in Indonesia are the poor people percentage aged above 15 years old and unemployment, the people percentage aged above 15 years old and employed in the agricultural sector, the literacy rate of the poor aged between 15 to 55 years old, and the life expectancy rate. Keywords: fixed effect model, exponential adaptive kernel.
What problem does this paper attempt to address?