<![CDATA[Unemployment is crucial issue in South Kalimantan Province as it contribute to increased poverty and the slowing down of regional economic growth. Therefore, reliable forecasting method is needed to support the government in formulating strategic policies. This study aims to compare and identify a representative time series forecasting model for the TPT in South Kalimantan Province. This research compares the performance of the DES Holt 1-parameter, 2 parameter, and ARIMA (1,1,0) models using 19 years of semi-annual unemployment rate data (2005-2024). The models applied parameter optimization for and , obtained by trial and error and by using R Studio, while accuracy was evaluated using MAPE. The research results show that the DES Holt-2parameter model with optimal parameters and . Demonstrated better statistical performance compared to the DES Holt 1-parameter and ARIMA (1,1,0) models because it yielded the smallest MAPE error value of 4,66. However, the ARIMA (1,1,0) models is superior because it is more capable of capturing dynamic trend changes in the region, whether the changes are short-term of long-lasting, and can better reflect socioeconomic structural changes. The contribution of this research is to provide a scientific basis and a more superior and representative forecasting model, namely ARIMA (1,1,0), for the South Kalimantan Provincial government in formulating policies for job creation, improving labor productivity, and strengthening the regional economy in a sustainable manner.]]>
