<![CDATA[The importance of information about the weather, especially information about rainfall that is more precise is needed so that research is needed to estimate and predict the occurrence of rain. There have been many studies reporting how to identify rainfall forecasts, but they still have limitations, especially in identifying rainfall in detail in each province in Indonesia. In this study, forecasting rainfall in Kendari City was carried out using two different time series methods, namely the trend method and the exponential smoothing method. The use of these two different methods was carried out with the aim of comparing the accuracy of the two models in predicting time series data, especially rainfall data in Kendari City. This research was conducted from January 2023 to June 2023, using rainfall data from Kendari City. In this study, the following conclusions were obtained: 1) the volume of monthly rainfall in Kendari City fluctuated, which spread in the range of 0 – 578.60 mm with an average of 228.26 mm per month, where on average the highest rainfall occurred in June and the lowest rainfall occurred in October; 2) the Trend method, whether it's Linear Trend, Quadratic Trend or Exponential Growth Trend, is not appropriate for estimating and forecasting Kendari City's monthly rainfall because the resulting estimate has quite high residuals with a relatively low level of accuracy; 3) the Winter's Exponential Smoothing method is quite good for estimating and forecasting Kendari City's monthly rainfall because the resulting estimate has quite low residuals, with a relatively high degree of accuracy; and 4) based on the RMSE, MAE and BIC criteria, it shows that the Winter's Exponential Smoothing model with an alpha (level) parameter value of 0.85, a gamma (trend) of 0.05 and a delta (seasonal) of 0.001 provides a better level of estimation and forecasting accuracy when compared to the Trend model, both Linear Trend and Quadratic Trend. Keywords: Rainfall, exponential smoothing, trend method, time series]]>
