Time series analysis is an effective statistical methodology for modelling time series data (a series of observations collected over time) and forecasting future observations in many areas, including economics, social sciences, physical and environmental sciences, medicine, and signal processing. This course presents the fundamental principles of time series analysis including mathematical modeling of time series data and methods for statistical inference. Topics covered will include modeling and inference in the following models : smoothing methods, decomposition methods, (nonseasonal/seasonal) autoregressive moving average (ARMA) models, unit root and differencing, (nonseasonal/seasonal) autoregressive integrated moving average (ARIMA) models, spectral analysis, (generalized) autoregressive conditionally heteroscedastic models, time regression models with autocorrelated error, lagged regression, and vector autoregressive (VAR) model.
Open to MATH, STATISTC, BIOSTATS, and COMPSCI graduate students. Note: Prerequisites: STAT 607/608 or equivalent for familiarity with maximum likelihood estimation. STAT 625 or 705 or equivalent for familiarity with linear algebra, specifically in the context of regression, recommended but not required.
Permission is required for interchange registration during the add/drop period only.