<![CDATA[This study aims to analyze the factors influencing household poverty status in East Kalimantan by comparing three models: Weighted Logistic Regression (WLR), Naive Bayes, and Binomial Regression with a generalized logit link function (glogit). The data used were obtained from the National Socioeconomic Survey (SUSENAS) conducted in March 2023. Parameter estimation was performed using a Bayesian approach with the Hamiltonian Monte Carlo (HMC) algorithm through the RStan program. The analysis results indicate that the number of household members, place of residence, education level of household head, employment status of household head, age of household head, and dependency ratio are significant variables affecting household poverty status in East Kalimantan. The comparison of the three models' performance shows that the WLR and Naive Bayes models are better at detecting poor households compared to the Binomial Regression with a generalized logit link function model, despite the Binomial Regression model showing higher overall accuracy. These findings provide important insights into the determinants of poverty and the effectiveness of various models in handling unbalanced binary data.]]>
Copyrights © 2023
