In recent years, the weather and climate are unpredictable and the most visible is the rotation of the rainy season and the dry season. The extreme changes in rainfall can cause disasters and losses for the community. For that we need to predict the rainfall to anticipate the worst events. Rainfall is included in the periodic series data, so the forecasting method that can be used is the ARIMAX model which is ARIMA model expanded by adding the exogen variable. The aim of this research is to predict the rainfall data in Pangkalpinang City, Indonesia. The best model for each rainfall is ARIMAX (0,1,3) for monthly rainfall data and ARIMAX (0,1,2) for maximum daily rainfall. This research shows that there is an influence maximum wind speed variable to monthly rainfall and maximum daily rainfall in the Pangkalpinang City. Nevertheless, when viewed from the ARIMA and ARIMAX models based on the obtained AIC value, the ARIMAX value is still better than ARIMA. However, the prediction value using ARIMAX needs to increase again to take into account seasonal data rainfall. Then, possible to add other exogeneous factors besides maximum wind speed.

East Belitung Regency is one of the regencies located on Belitung Island. East Belitung Regency has a tropical and wet climate with a fairly high variation of rainfall. Rainfall forecasting is an important thing to model because of the many uses of rainfall forecasting results such as irrigation planning, flood prediction, erosion prediction and others. This study aims to predict rainfall for the next 5 years by using a time series model by reviewing the heteroscedasticity of the data. From the results of the analysis of rainfall in East Belitung Regency with a seasonal pattern. The best model used is ARIMA (0, l, l)(2, l, l)12 with insignificant heteroscedasticity.