Regression analysis for yield loss of oil palm due to Ganoderma disease

摘要

Oil palm industry is well-established in Malaysia as well in Indonesia. The industry is facing a devastating crop disease which is Ganoderma Basal Stem Rot Disease (BSR) or popularly known as Ganoderma disease caused by a fungal called as Ganoderma boninense. The objective of this study is to develop a yield loss model of oil palm due to the disease. Backward elimination based regression method was used in developing the yield loss model where the independent variables also include interaction categorical variables up to the second order of interaction. Residual analysis was conducted on the best model developed and it showed that the mean value and standard deviation of the standardized residuals were zero and one respectively. The distribution of the standardized residual was also normal, homoscedastic, no presence of outlier. The value of the mean absolute percentage error (MAPE) which was used to measure the forecast performance of the best model was considered reasonable forecasting. The yield loss model developed in this study can be used in estimating the economic loss as well as the economic advantage of any Ganoderma disease control.