<![CDATA[General Background Local tax revenue plays a crucial role in supporting regional development and fiscal sustainability. Specific Background Sidoarjo Regency experienced a general increase in local tax revenue from 2014 to 2024, with a significant decline in 2020 due to the COVID-19 pandemic. Knowledge Gap Accurate forecasting methods are required to anticipate future revenue and support regional financial planning, yet the most suitable model for Sidoarjo’s tax data remains unclear. Aims This study aims to forecast Sidoarjo local tax revenue for 2025 and determine the most appropriate Exponential Smoothing model. Results Single, Double, and Triple Exponential Smoothing methods were applied and compared using Mean Absolute Percentage Error. The Triple Exponential Smoothing model produced the lowest MAPE value of 8.37% and a forecasted revenue of Rp. 1,507,007,282,941 for 2025. Novelty This study provides a comparative application of Exponential Smoothing models using long-term local tax data with explicit accuracy evaluation. Implications The findings support evidence-based regional revenue planning and contribute to methodological selection in public sector financial forecasting. Highlights: Comparative evaluation of three Exponential Smoothing models Triple Exponential Smoothing yields the lowest forecasting error Local tax revenue trend shows post-pandemic recovery Keywords: Forecasting, Local Tax Revenue, Exponential Smoothing, Time Series Analysis, Regional Finance]]>
