Poverty Line Forecasting Model in North Sumatera Province Using Double Exponential Smoothing Method

One of the information that can be used as a consideration in preparing poverty alleviation strategies is data on the existence of poverty rates in the future. The poverty line is a variable for calculating the number of poor people. This study aims to determine the model and predict the poverty line in North Sumatra Province using the Double Exponential Smoothing method from Holt. This study uses time series data and predicts the poverty line for the coming year. In this study, data pattern analysis was carried out where the data pattern shows a trend, which means that the Double Exponential Smoothing Method from Holt is the appropriate method to use. Then the best parameter value was determined where the parameters used were the Alpha (α)and Gamma (γ)with the smallest MAPE (Mean Absolute Percentage Error) value. With the trial and error method, the Alpha parameter value ( α)of 0.1 and gamma ( γ)of 0.1 and the MAPE value of 2 percent) were obtained. The results of the study showed that this forecasting model has very good performance and the poverty line value continued to increase in the coming year.