TELBS Regression Analysis to Determine the Influence of Coffee Land on Coffee Production in Indonesia in 2023 Using Python Programming Language

Abstract

Indonesia, as one of the world's largest coffee producers, is renowned for its diverse range of high-quality coffees such as Arabica, Robusta, and Liberica. Coffee production is influenced by various factors, including the extent of plantation land. Coffee production data may contain outliers due to factors like weather changes, pest attacks, inconsistent farming practices, or recording errors. These challenges can be addressed using robust regression methods, with one such estimation being Tabatabai Eby Li Bae Singh (TELBS) estimation. TELBS estimates model parameters by minimizing an objective function. In this study, a TELBS estimation model was applied to Indonesian coffee production data in 2023, with the dependent variable being coffee production quantity and the independent variable being plantation land area. Parameter testing using t-tests indicated that plantation land area significantly influences coffee production in that year at a significance level of 0.05. The TELBS estimation model yielded a coefficient of determination of 96.51%, demonstrating its capability to explain a substantial portion of the data's variance.