Komparasi Generalized Poisson Regression dan Negative Binomial Regression pada Pemodelan Jumlah Curah Hujan

Abstract

The study aims to compare Generalized Poisson Regression (GPR) and Negative Binomial Regression (NBR) in modeling rainfall amounts based on meteorological factors in South Sumatra from 2021 to 2023, calculated from January to December. Meteorological data such as air temperature, humidity, wind speed, and air pressure were collected for analysis. The analysis was conducted using R Studio. The results indicate that GPR was used to address overdispersion, with significant variables being X1 and X1 , while NBR showed significant influence from variables X1, X1, and X1 on rainfall amounts. The NBR model provided the best fit with the lowest AIC value (427.97).