Time Series

Catalog description

This course introduces the statistical methodology and models required to analyze time series data in practice. The course emphasizes both modeling methodology (model identification, estimation and diagnostics) and the practical implementation of time series modeling using existing statistical software. Topics include Analysis of time series and forecasting methods, Stationary processes, ARMA models, Autocorrelation function, Spectral analysis, Non stationary time series, ARIMA models, SARIMA models, Unit roots and Volatility models. Typically offered Fall Spring. Prerequisite(s): STAT 34600 FOR LEVEL UG WITH MIN. GRADE OF C AND STAT 43100 FOR LEVEL UG WITH MIN. GRADE OF C Course Learning Outcomes 1. Define time series data in an appropriate statistical framework. 2. Summarize and carry out exploratory and descriptive analysis of time series data. 3. Model univariate time series data with Autoregressive and Moving Average Models. 4. Demonstrate a working knowledge of sampling techniques. 5. Model identification, model estimation, and assessment of the suitability of the model. 6. Use a model for forecasting and determining prediction intervals. 7. Use relationships between time series variables cross correlation and regression models. 8. Analyze the frequency domain - Periodograms, Spectral density, identifying the important periodic components of a series. 9. Explore some volatility models. View Class Schedule