Generalized Linear Mixed Model Tree (GLMM-Tree) for The Classification of Direct Cash Transfer Recipients in West Java Province

Nawawi, M. Ichsan

Jurnal Matematika dan Statistika serta Aplikasinya (Jurnal MSA)
Vol 13 No 2 (2025): VOLUME 13 NO 2, 2025

Generalized Linear Mixed Model Tree (GLMM-Tree) is a statistical method that combines the concepts of decision tree and Generalized linear mixed model (GLMM). Here are some key advantages including Flexibility in Handling Different Types of Data, Incorporation of Random Effects, Handling of Non-linear Relationships, Interpretability, Variable Selection, Robustness to Outliers, Capturing Interactions, No Need for Parametric Assumptions. The purpose of this study is to compare the GLMM and GLMM-tree methods for the classification of direct cash transfer recipients in West Java with 25890 observations using the GLMM-tree method. Looking at the MSE and RMSE values, GLMM-tree is superior to GLMM for both training and testing data