Analysis of the Effect of Climate Variability Risk on Rice Farming Productivity Using Robust Regression

AbstractThis study aims to analyze the effect of climate variability risk on rice farming productivity and terrestrial riceproduction. In this study, the data analysis method was carried out using multiple linear regressionmodels, and to estimate the parameters is done using Ordinary Least Square (OLS). Because of theproductivity of rice farming and terrestrial rice production under study there are outlier data, so we need theright method to perform data analysis. Robust Regression is an appropriate method for dealing withirregularities caused by outliers. Furthermore, to measure the level of risk, it is done using Conditional Valueat-Risk (CVaR). Based on the analysis results show that wind speed, maximum temperature andminimum temperature have a significant effect on rice farming, while rainfall does not have asignificant effect. The robust regression estimator has a determination value of 0.991 which means it has avery strong correlation. The effect of climate variables on terrestrial rice production shows that wind speed,minimum temperature, and rainfall have a significant effect on terrestrial rice production, whilemaximum temperature has no significant effect. The robust regression estimator in this case gives adetermination value of 0.574 which means it is quite strong. The level of losses due to damage due to climatevariables is estimated to reach a minimum CVaR value of 58.4459 tons/ha at the 0.96 confidence level.