STAT 5820 - Time Series Analysis

The development and practical use of seasonal and non-seasonal ARIMA (Autoregressive Integrated Moving Average) Box-Jenkins time series models is presented. Identification of correct time series models, estimation of model parameters, and diagnostic checks of identified models will be covered. The uses of these models for forecasting future trends and assessing interventions will be examined. Extensive data analysis using SAS, MINITAB, and Splus/R statistical packages are included. Topics include: regression time series models, autocorrelation, partial autocorrelation, Yule-Walker equations, differencing, stationarity, autocorrelation models, moving average models, seasonality, invertibility, and Box-Pierce tests.

Prerequisites/Corequisites: Prerequisites: STAT 3640 and STAT 5680.

Credits: 3 hours

Notes: Open to Upperclass and Graduate students.

Print-Friendly Page (opens a new window)