Robust Regression Analysis Of Gm Estimation On The Poverty Gap Index Of Indonesian Provinces

Poverty has been a severe problem in Indonesia since the post-independence era until today. One indicator that can be used to measure the poverty level in a region is the poverty gap index, which describes the average size of the gap between each population and the poverty line. The data on the poverty gap index in Indonesia in 2022 contains outliers and misleading data that are not normally distributed, so the least squares method is inappropriate. One method that can be used to overcome the outlier problem is robust regression analysis. This study aims to determine the Generalized M (GM) estimation robust regression model and the factors that affect the poverty gap index in provinces in Indonesia. The estimation used is GM estimation, the development of M estimation when M estimation is less sensitive to outliers. The results showed that the GM estimation robust regression model has a value of 100% It was also found that the factors that significantly affect the poverty gap index in Indonesia in 2022 are the percentage of poor people, the Gini ratio, the poverty line, and the percentage of households with a PLN electricity source.