@article{Aliu_Zubedi_Yahya_Oroh_2022, title={The Comparison of Kernel Weighting Functions in Geographically Weighted Logistic Regression in Modeling Poverty in Indonesia}, volume={18}, url={https://journal.unhas.ac.id/index.php/jmsk/article/view/19567}, DOI={10.20956/j.v18i3.19567}, abstractNote={<p>Indonesia is a developing country that is facing poverty. The percentage of the poor population in Indonesia in 2020 increased by 0.97 percent from 2019. A suitable analysis to overcome poverty in Indonesia is the regional effect, namely Geographically Weighted Logistic Regression (GWLR). This study aimed to compare the weighting functions of the Fixed Gaussian Kernel, Fixed Tricube Kernel, and Fixed Bisquare Kernel in the GWLR model in modeling poverty in Indonesia in 2020. The best model can determine significant factors that affected poverty in Indonesia in 2020. This study used the percentage data of poor population  and the factors affecting it, namely the Open Unemployment Rate , Human Development Index , and Total Population  in 34 Provinces in Indonesia. This study indicates that the GWLR model with the Fixed Gaussian Kernel weighting function is the best in modeling poverty in Indonesia in 2020 based on the smallest Akaike Information Criterion Corrected (AlCc) value. The GWLR model with the Fixed Gaussian Kernel weighting function shows the Open Unemployment Rate as a significant factor affecting poverty in Indonesia in 2020 in 10 provinces in Indonesia, namely Aceh, North Sumatra, West Sumatra, Riau, Jambi, South Sumatra, Bengkulu, Lampung, DKI Jakarta, and Banten.</p> <p> </p>}, number={3}, journal={Jurnal Matematika, Statistika dan Komputasi}, author={Aliu, Muftih Alwi and Zubedi, Fahrezal and Yahya, Lailany and Oroh, Franky Alfrits}, year={2022}, month={May}, pages={362-384} }