This course is about forecasting and some of the statistical techniques that can be used to produce forecasts. Main purpose of this course is to present forecasting principles and applications in a comprehensive way. Since the emphasis is on the application of the forecasting techniques and all of the applications are made by EViews statistical software, lectures are given in computer lab. So are examinations. Questions of examinations are selected so that at least seventy percent of them have to be answered by using EViews.

EViews Basics: Introduction to EViews, workfile basics, sample, using expressions, working with series, scalars and graphs, data objects, importing data, exporting data. Time series and cross sectional data, graphical summaries, time plots and time series patterns: trend, cyclical, seasonal and irregular pattern. Numerical summaries: univariate statistics, bivariate statistics. Measuring forecast accuracy: mean error, mean absolute error, mean squared error, root mean squared error, mean percentage error, mean absolute percentage error and Theil's U statistic. Transformations and calendar adjustments: mathematical (square root, cube root, negative reciprocal and logarithmic) transformations and calendar (month length and trading day) adjustments. Moving averages: simple, centered, double and weighted moving averages. Time series decomposition: classical additive decomposition, classical multiplicative decomposition, finding and interpreting seasonal index, deseasonalizing the data, forecasting with decomposition methods. Exponential smoothing methods: simple exponential smoothing, Holt's exponential smoothing, Winters' exponential smoothing. Introduction to forecasting with regression methods: simple regression, the least squares estimation, the correlation coefficient, simple regression and the correlation coefficient, forecasting using the simple regression model.