BIOSTAT645: Time Series Analysis with Biomedical Applications
Graduate level
Residential
Fall term(s) for residential students;
3 credit hour(s) for residential students;
Instructor(s): Staff (Residential);
Not offered 2024-2025
Prerequisites: Biostat 602, Biostat 650 or Perm. Instr
Description: Introduction to statistical time series analysis with an emphasis on frequency domain (spectral) methods and their applications to biomedical problems. Topics include autocorrelation, stationarity, autoregressive and moving average processes, power spectra, periodgrams, spectral estimation, linear filters, complex demodulation, autoregressive integrated moving average (ARIMA) models, cross-correlation, cross-spectra, coherence, time and frequency domain linear regression. The methods will be illustrated in applications to various areas of public health and medical research such as environmental health, electrophysiology, and endocrinology.
