M568 : Time Series Analysis
This course provides an introduction to the theory and practice of Time Series Analysis, the field of Statistics involved with the analysis of dependent data that are ordered in time. This course gives approximately equal weight to "time domain'' analysis and "frequency (spectral) domain'' analysis. It deals primarily with the usual discrete-time univariate time series, but bivariate and continuous-time processes are treated briefly.
Capacity: Never been a problem for this course.
Instructor: Richard Bradley - email@example.com
Other Contact(s): Richard Bradley - firstname.lastname@example.org
Sequence: No. It is a one-semester course.
Prerequisites: MATH M466 or STAT S420 or consent of the instructor. The main prerequisite is a sound familiarity, at the undergraduate level, with basic real analysis, probability,and statistics, and the elementary properties of complex numbers (up to and including the form exp(it) = (cos t) + i(sin t) ).
Algebra Required?: Used for notations. Used in homework.
Calculus Required?: Used for concepts. Used in homework.
Day(s) per week offered: Three lectures per week.
Substantive Orientation: Time Series Analysis is used in numerous fields, including Economics, Engineering, Astronomy, Geology, among others
Statistical Orientation: Frequentist.
Books used: The main text is C. Chatfield, The Analysis of Time Series: An Introduction. Supplementary sources: G.E.P. Box and G.M. Jenkins, Time Series Analysis: Forecasting and Control; P. Bloomfield, Fourier Analysis of Time Series, An Introduction; and P.J. Brockwell and R.A. Davis, Time Series: Theory and Methods.
Applied/Theoretical: Slightly more theoretical than applied.
How the software is used: Data analysis
Problem Sets: six written problem sets
Data Analysis: Three computer assignments
Exams: A Final Exam
Keywords: The analysis of dependent data ordered in time