|
Course Webpage:
http://www.nes.ru/~sanatoly/AE/AE.htm
Period
3, 2000/01
Instructor:
Stanislav
Anatolyev
The course
is devoted to the modern applied time series analysis. We review
used-to-be-popular linear nonstructural models like ARIMAs,
study VARs, and then turn to nonlinear models of the mean, like
sample splitting and chaos, and of the variance, like ARCH.
If time permits, structural models with Rational Expectations
and bootstrap methods in time series will be discussed. We focus
on methods, on the one hand, and relevant applications, on the
other. The course presumes intensive use of classical publications
in the field and computer work.
TENTATIVE
SYLLABUS
I. Univariate
time series: modeling the mean (2 weeks)
- Stationary
ARMA models: properties, estimation, analysis and forecasting.
- Nonstationary
univariate time series: stochastic and deterministic trends,
unit root tests
- Nonlinear
time series modeling of the mean: thresholds, structural
breaks, chaos
II. Multivariate
time series: modeling the mean (2 weeks)
- Stationary
VAR models: properties, estimation, analysis and forecasting.
- Nonstationary
multivariate time series: spurious regression, cointegration,
common trends
- Modeling
the variance (2 weeks)
- The
class of ARCH models: properties, estimation, analysis and
forecasting.
- Time-varying
risk and ARCH-in-mean
- Stochastic
volatility models
IV. Other
aspects of time series analysis (1 week)
- Structural
econometric modeling and GMM
- Bootstrapping
in time series: parametric bootstrap, block bootstrap, sieve
bootstrap, grid bootstrap.
TEXTS
Hamilton,
James. Time Series Analysis, Princeton University Press
Cochrane,
John. Time Series for Macroeconomics and Finance, Lecture
Notes, University of Chicago, Graduate School of Business
Mills,
Terence. The Econometric Modelling of Financial Time Series,
Cambridge University Press
Taylor,
Stephen. Modelling Financial Time Series, John Wiley
& Sons
Harvey,
Andrew. Time Series Models, MIT Press
|
