CD Skripsi
Penerapan Metode Exponential Smoothing Untuk Peramalan Angka Inflasi Kota Pekanbaru
ABSTRACT
Forecasting using the exponential smoothing method is a time series data forecasting. In this study the data used for analysis is inflation data. Based on the pattern of time series data, the data obtained then analyzed with single exponential method and triple exponential method consisting of Holt-Winters multiplicative and Holt-Winters additive. Measeurement of forecasting accuracy for selecting the best model can be done by selecting the lowest forecasting error value using the Mean Absolute Percentage Error (MAPE). This study shows that the single exponential smoothing method with a smoothing constant alpha = 0,3 has the minimum MAPE value compared to the triple exponential smoothing method. Forecasting results using the exponential smoothing method with smoothing constant alpha value = 0,3 shows that the inflation rate of Pekanbaru City has decreased for the coming period.
Keywords: Time Series, inflation, forecasting, single exponential smoothing method, triple exponential smoothing method, MAPE.
Tidak tersedia versi lain