<![CDATA[This study aims to assess and compare the performance of three forecasting models—Autoregressive Integrated Moving Average (ARIMA), Seasonal Autoregressive Integrated Moving Average (SARIMA), and Gated Recurrent Unit (GRU)—in predicting car sales in Indonesia. The dataset presents intricate seasonal patterns and non-linear fluctuations, which pose challenges for conventional statistical models. The ARIMA model, suited for linear and stationary data, struggled to capture the complexities of the sales trends. While SARIMA, an enhanced version of ARIMA, aimed to handle seasonal components, it also failed to provide accurate predictions. In contrast, the GRU model, a deep learning-based technique, exhibited the best results in terms of predictive accuracy, with significantly lower values for Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). The results highlight the superior performance of the Gated Recurrent Unit (GRU) model in forecasting car sales. This superiority is reflected in lower error values across all evaluation metrics compared to ARIMA and SARIMA. The GRU model provides accurate forecasts for complex business decision-making]]>
