CSI 996
Unit Roots and Cointegration
Fall, 2015
Instructor:
James Gentle
Reading Course
This course covers various time series models, in particular those with unit root
nonstationarity. Both VAR and SVAR models are considered, and error-correction
models related to both are developed. The effects of unit root processes are
explored and various tests for unit roots are considered and compared.
Spurious regressions in cointegration are studied. The Engle-Granger two-step
development of an error-correction model is studied, as well as its extension
to a structural form
Text
The text for the course is
Pfaff, Bernhard (2008), Analysis of Integrated and Cointegrated Time
Series with R, second edition, Springer, New York.
Software
We will use R and various R packages, including primarily urca, but also possibly
tseries, fArma, fracdiff, fUnitRoots, and uroot.
Topics and Assignments
Read through the entire text.
Exercises are from Pfaff (2008).
- VARs and SVARs: page 51, Exercises 7, 8, 9
- Long memory processes: page 71, Exercises 4, 5, 6
- Error-correcting models: page 127, Exercise 2, and page 159, Exercise 3