Colleges

Course Description: Introduction - General Trend - Seasonality - Cyclicality. Markov Chains - Data Transformation Methods: Differencing Method - Seasonal Adjustment. Forecasting: How to Forecast the Future - Possible Accuracy of Forecasting - Forecasting Using Regression versus Time Series Forecasting. Some Useful Numerical Measures for Forecasting Accuracy: MAD, MSE, MAPE. Time Series Smoothing and Decomposition Methods: Averaging Smoothing Methods - Exponential Smoothing Methods: Simple Exponential Smoothing - Weighted Exponential Smoothing - Decomposition Methods: Trend Component Fitting - Various Types of Moving Averages. Box-Jenkins ARIMA(p,d,q) Models for Time Series Forecasting: Identifying the Characteristics of These Models and Their Autocorrelation and Partial Autocorrelation Functions. Dealing with Data with Seasonal Components - Fitting These Models to Computer-Generated Data and Real Data. Diagnostic Tests on Model Residuals and Model Selection Criteria. Practical Exercises in Data Analysis and Writing Comprehensive Reports Using Ready-Made Statistical Software.