TY - JOUR T1 - Comparison of ARIMA, Fuzzy Regression and Exponential Smoothing methods in the value added of the industrial sector forecasting TT - مقایسه الگوی خود توضیح جمعی میانگین متحرک، روش‌های هموارسازی و رگرسیون فازی در پیش‌ بینی ارزش‌افزوده بخش صنعت ایران JF - raja-tfe JO - raja-tfe VL - 2 IS - 3 UR - http://tfe.raja.ac.ir/article-1-46-en.html Y1 - 2014 SP - 37 EP - 54 KW - Industry KW - Value Added KW - Forecasting N2 - Nowadays, forecast economic variables play an important role in economic planning and policy making and a variety of methods to forecasting economic variables are used. Although it may be accurate in some cases is not important in predicting, but obviously many variables to predict the short-term policy is important.Due to the importance of industry and its share in GDP, in order to predict the value added of the industrial sector in Iran during the period 1340-1389, in this paper four methods have been evaluated: Auto Regressive Integrated Moving Average (ARIMA), Fuzzy Regression (FR), Single Exponential Smoothing with Trend (SEST) and Double Exponential Smoothing with Trend (DEST). Comparing the accuracy of predictions, based on two criteria RMSE and R2 indicated that single exponential smoothing with trend (SEST) and fuzzy regression (FR) had the best results in forecasting the value added of the industrial sector in Iran. M3 ER -