The Analysis of Time Series (4)
Classes: MWF 1300-1400 in 330 Evans
Lab Section: Fri 1100-1300 in 330 Evans
Meant for graduate students in Statistics and other Departments.
Initially the course will cover the conventional statistical approaches to time series in both the time and frequency domains, including state space. Then it will move to nonlinear time series and systems methods currently appearing in the applied mathematics and physics literatures. There will be comparative discussion of the approaches.
The two recommended texts are:
Kantz, H. and Schreiber, T. (1997). Nonlinear Time Series Analysis. Cambridge University Press.
Quoting from this work, "... the concept of a phase space representation rather than a time or frequency domain approach is the hallmark on nonlinear dynamical time series analysis."
and
Brillinger, D. R. (2001). Time Series Data Analysis and Theory. SIAM Reprint.
Text for lab:W.N. Venables and B.D. Ripley Modern Applied Statistics with S-Plus. Springer. The computer package TISEAN will also be used.
Prerequisites: Statistics 101&102 or 134&135 or equivalents.
Grading: The grade will come from weekly problems, a midterm and a Project involving the analysis of time series data.
Instructor: David R. Brillinger
Office Hours:M W 4-5. In 417 Evans.
GSI: Michael Last
Office Hours:Th 3:30-5. In 321 Evans.
Course homepage: http://www.stat.berkeley.edu/~brill/Stat248/index.html