MODELING THE NUMBER OF POOR POPULATION IN EAST JAVA USING QUANTILE REGRESSION

Abstract

The economic development of East Java continues to increase every year. However, this increase is not directly proportional to a significant decrease in poverty rates. Therefore, research is needed to determine the factors influencing poverty in East Java. This is important because it can be used as a consideration for the East Java Provincial Government in designing strategies to reduce poverty. In the case of the number of poor people in East Java, there are outlier data, so the quantile regression method is used to overcome this. This study uses several quantile values, namely 0.25, 0.50 and 0.75. Based on the results of the quantile regression parameter estimation, one significant category at all quantile levels is the Average Length of Schooling variable. From the quantile regression model, four categories of Poor Population are obtained: low, medium, high, and very high. Based on the classification of the Poor Population in East Java in 2023, there are four districts/cities with a low number of poor people, 18 districts/cities with a moderate number of poor people, and 16 districts/cities with a high number of poor people.