Pendekatan Non Linier pada Estimasi Produktivitas Kelapa Sawit di Kabupaten Tapanuli Selatan

Abstract

Palm oil productivity is a crucial aspect of the plantation industry in Indonesia, contributing significantly to both regional and national economies. Changes in climate conditions and unsustainable farming practices can directly impact the productivity and quality of palm oil, underscoring the need for thorough analysis. This study aims to estimate palm oil productivity in Tapanuli Selatan District using quadratic and exponential regression models to determine the superior model.The conclusion of this research indicates that the quadratic regression model provides more accurate estimations of palm oil production with an R² value of 0.971, demonstrating its ability to explain 97.1% of the variability in palm oil production. Both models show that land area and rainfall significantly influence palm oil production, but the quadratic model offers more precise and detailed estimations. Therefore, for more accurate and reliable palm oil production estimates, the quadratic regression model is the preferred choice, as it captures the complex effects of land area and rainfall on production more effectively compared to the exponential regression model.