E3S Web Conf.
Volume 73, 2018The 3rd International Conference on Energy, Environmental and Information System (ICENIS 2018)
|Number of page(s)||4|
|Section||Health, Safety and Environment Information Systems|
|Published online||21 December 2018|
The Step Construction of Geographically Weighted Panel Regression in Air Polluter Standard Index (APSI) Data
Department of Statistics, Faculty of Science and Mathematics, Diponegoro University, Semarang - Indonesia
* Corresponding author: firstname.lastname@example.org
Geographically Weighted Panel Regression or GWPR is a local linear regression model that combines GWR model and panel data regression model with considering spatial effect, especially spatial heterogeneity problem. This article is focused on the soft computation of GWPR model using Fixed Effect Model (FEM). Parameter estimation in GWPR is obtain by Weighted Least Squares (WLS) methods and the resulting model for each location will be different from one to another. This study will compare the fixed-effect GWPR model with several weighting functions. The best model is determined based on the biggest coefficient of determination (R2) value. In this study, the model is applied in the Air Polluter Standard Index (APSI) in Surabaya City, East Java. The results of this study indicate that Fixed Effect GWPR model with a fixed exponential kernel weighting function is the best model to describe the APSI because it has the smallest AIC.
Key words: APSI / Fixed Effect, GWPR / Panel Regression
© The Authors, published by EDP Sciences, 2018
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.