GEOGRAPHICALLY WEIGHTED PANEL LOGISTIC REGRESSION SEMIPARAMETRIC MODELING ON POVERTY PROBLEM

Aliyah Husnun, Azizah and Adji Achmad Rinaldo, Fernandes and Rosita, Hamdan (2023) GEOGRAPHICALLY WEIGHTED PANEL LOGISTIC REGRESSION SEMIPARAMETRIC MODELING ON POVERTY PROBLEM. MEDIA STATISTIKA, 16 (1). pp. 47-58. ISSN 1979 – 3693

PDF Geographycally weighted panel logistic regression.pdf Download (298kB)

Abstract

Regression analysis is a statistical method used to investigate and model the relationship between variables. Furthermore, a regression analysis was developed that involved spatial aspects, namely Geographically Weighted Regression (GWR). GWR modeling consists of various types, one of which is Geographically Weighted Logistic Regression Semiparametric (GWLRS), an extension of the Logistic GWR model that produces local and global parameter estimators. In this study, it is proposed to combine the GWLRS model using panel data or Geographically Weighted Panel Logistic Regression Semiparametric (GWPLRS). The case study used in this research is the problem of poverty in 38 regions/cities in East Java, Indonesia, in 2018 – 2022 as seen from the Poverty Gap Index. The weights used in this research are the adaptive gaussian kernel weighting functions. The results of the parameter significance test show that the Human Development Index as global variable has a significant effect on each region/city.

Actions (For repository members only: login required)

View Item