METODE STEVENSON PORTER FUZZY TIME SERIES DAN PEMULUSAN EKSPONENSIAL UNTUK PROYEKSI DATA RUNTUN WAKTU STUDI KASUS : DATA PRODUK DOMESTIK REGIONAL BRUTO (PDRB) PROVINSI KEPULAUAN BANGKA BELITUNG; STEVENSON PORTER OF FUZZY TIME SERIES AND EXPONENTIAL SMOOTHING METHODS FOR TIME SERIES PROJECTION CASE STUDY: GROSS DOMESTIC REGIONAL PRODUCT (GDRP) OF BANGKA BELITUNG ISLANDS PROVINCE

Dalimunthe, Desy Yuliana, Gunardi

2015 | Disertasi | FMIPA

In this paper, process forecasting of Gross Domestic Regional Product (GDRP) is using Stevenson Porter fuzzy time series and exponential smoothing methods by current prices, this means that the calculation of value added goods and services using the price in the current year. Based on the forecast result of both methods campared by see the Mean Square Error (MSE) and Mean Absolute Percentage Error (MAPE) value on each method. GDRP data is used in this paper start from the first quarter of 2007 until the period of quartal II of 2014 with MSE value from fuzzy time series method is 87.680.287 with MAPE value is 3.885955x10-6 and the percentage error is 0.48, in the other hand MSE value from exponential smoothing method is 28.145.950.970 and MAPE value from this method is 8.10248x10-5 and the percentage error is 1.75. This result shows that Stevenson Porter fuzzy time series method has a higher accuracy rate than exponential smoothing.

Kata Kunci : forecasting; GDRP; Stevenson Porter fuzzy time series, Exponential Smoothing