<![CDATA[This research aims to forecast the profitability of Indonesian rural banks. The forecasting methods employed dynamic and static forecasting. Dynamic forecasting was represented by regression with Autoregressive Integrated Moving Average (ARIMA) errors while static forecasting was represented by Holt-Winters seasonality and Seasonal Autoregressive Integrated Moving Average (SARIMA). The regression with ARIMA errors included additional independent variables namely inflation and interest rate. The dependent variable being forecast was return on assets (ROA) of the rural banks. The data extended from January 2010 to July 2021. The data will be divided into training and test data. Training data extended from January 2010 to December 2020. The test data extended from January 2021 to July 2021. Training data were used to derive the models. The models generated then were used to yield forecasts for January until July 2021. The forecasts were later compared to the test data for accuracy. The research found that regression with ARIMA errors had the best forecast accuracy, followed by SARIMA and Holt-Winters seasonality. Therefore, the research proposed that regulators, analysts and all stakeholders of the Indonesian rural banks employ regression with ARIMA errors to predict the profitability and financial position of the rural banks.]]>
Copyrights © 2022
