PENERAPAN MODEL REGRESI PANEL KOMPONEN DUA ARAH PADA POLA CURAH HUJAN PROVINSI KALIMANTAN TENGAH

Abstract

Rainfall is the climate element that is most closely related to supporting the life processes of Indonesian people such as agricultural production, plantations, fisheries, and aviation. In 2014-2016, Indonesia experienced a drought due to a global climate anomaly called the El Nino phenomenon, where annual rainfall at that time tended to decrease from other years. While in 2020-2022, rainfall in Indonesia tends to increase from other years, this event is called the La Nina phenomenon. This study aims to describe the rainfall patterns that occur in each phenomenon and analyze the regression model of rainfall panels in Central Kalimantan province with a two-way component approach. Random Effects Model (REM) is the most appropriate model to be used in the phenomenon of La Nina. Fixed Effect Model (FEM) is the most appropriate model to be used in the El Nino phenomenon. Feasible Generalized Least Square is a parameter estimation method that is focused and used to estimate regression parameters in this study. Based on the results of regression analysis of panel data, for the phenomenon of La Nina obtained R2 value of 51.66% and found that the average air temperature variable tested significant. For the El Nino phenomenon, the value of R2 is 75.35% and it is found that there are no significant independent variables tested. Therefore, it can be expected that the increase in average air temperature can decrease the average rainfall value when the La Nina phenomenon occurs in Central Kalimantan province.