| dc.description.abstract | This study aims to forecast food price data for beef and chicken eggs in traditional markets of North Sumatra by applying the Holt-Winters method. The Holt-Winters Additive and Multiplicative methods will apply to food prices of beef and chicken eggs in the traditional markets of North Sumatra by determining the initial smoothing values of level, trend, and seasonal. The determination of the optimum parameter values α, β, and γ for each food price data of beef and chicken eggs in the traditional markets of North Sumatra. After obtaining the initial smoothing values for level, trend, and seasonal, and optimum parameter values, the smoothing values for level, trend, and seasonal are calculated. Then forecasting food price data for beef and chicken eggs in traditional markets of North Sumatra. After the forecast is obtained, the level of accuracy is calculated using RMSE and MAPE. Based on the results of the application of the Holt-Winters method, it can be concluded that the Holt-Winters method has a very good forecasting performance value. The MAPE accuracy level is <10% for all forecasting calculations, both Holt-Winters Additive and Holt-Winters Multiplicative. The Holt-Winters Multiplicative on beef price data is better than the Holt-Winters Additive because of a lower level of accuracy, namely RMSE = 1208,580 and MAPE = 0.680%. The Holt-Winters Additive for chicken egg price data is better than the Holt-Winters Multiplicative because of the lower accuracy, namely RMSE = 785,212 and MAPE = 2.459%. | en_US |
