Time Series Analysis and Applications
Listed in: Mathematics and Statistics, as MATH-25
Shu-Min Liao (Section 01)
Many real world applications deal with a series of observations collected over time. Some familiar examples are daily stock market quotations in finance, monthly unemployment rates in economics, yearly birth rates in social science, global warming trends in environmental studies, seismic recordings in geophysics, and magnetic resonance imaging of brain waves in medicine. In this applied course, students will learn how to model the patterns in historical values of the variable(s), as well as how to use statistical methods to forecast future observations. Topics covered will include time series regression, autoregressive integrated moving average (ARIMA) models, transfer function models, state-space models and spectral analysis. If time permits, additional topics will include autoregressive conditionally heteroscedastic (ARCH) models, Kalman filtering and smoothing, and signal extraction and forecasting. Students will get practice with various applications using statistical software. Four class hours per week.
Requisite: Mathematics 17 or 29 or consent of the instructor. Limited to 20 students. Spring semester. Professor Liao.
If Overenrolled: Priority will be given to Math majors.
Offerings2015-16: Offered in Spring 2016
Other years: Offered in Spring 2011