Introductory Time Series Analysis - STA 312

Introduces the basic concepts of regression analysis starting with two variable models then proceeds to three variable and multi-variable regression models. Thorough discussion of: The assumptions underlying linear regression models; Diagnostic tests, and correction methods for heteroscedasticity ,multicollinearity and serial correlations . The second part of the course introduces deterministic and stochastic time series models and discusses: Basic smoothing and extrapolation techniques; Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) for the different models; Stationarity, nonstationarity and Invertibility conditions; Model specification, Parameter estimation and forecasting for the different stationary time series models AR (p), MA (q), ARMA (p, q), and the homogenous non-stationary models of order d ARIMA. Prerequisite: STA 315 or MAT 326.