<![CDATA[This research is motivated by the significant role of rainfall as a natural occurrence that has a significant impact on various sectors. The main objective of this study is to evaluate the influence of exogenous variables in rainfall forecasting by comparing the accuracy of SARIMA, SARIMAX, hybrid SARIMA-ANN, and hybrid SARIMAX-ANN models. The initial modeling of this research did not involve exogenous variables and resulted in a MAPE (Mean Absolute Percentage Error) of 43,65% for the SARIMA model and 42,86% for the hybrid SARIMA-ANN model. The findings show that incorporating the exogenous variable of sunlight duration positively contributes to improving the accuracy of rainfall forecasting using the SARIMAX model with a MAPE of 28.69% and the hybrid SARIMAX-ANN model with a MAPE of 27,03%. This research contributes to the development of rainfall forecasting methods by incorporating other influencing variables.]]>
Copyrights © 2024
