Economics 512: Financial Econometrics
This is a research seminar course in financial econometrics. The focus will be on the statistical modeling of financial time series (asset prices and returns) with an emphasis on modeling volatility and correlation. The learning goals/objectives of the course are to (1) survey the relevant theoretical and empirical econometric literature; (2) introduce current research topics; (3) use statistical software to get hands-on experience with real world data. Topics to be covered include:
o Empirical properties and stylized facts
o Marke efficiency and predictability
� Volatility modeling
o Univariate and multivariate autoregressive conditional heteroskedasticity (ARCH) family of models
o Stochastic volatility models
o Applications to risk management and derivatives pricing
� Ultra high frequency time series
o Market microstructure models
o Realized variance, covariance and bi-power variation
o Applications to volatility modeling and market microstructure models
� Continuous time models
o Common models for equity and interest rates
o Applications to derivatives pricing
o Relationship to realized variance
Weekly homework assignments and computer lab work using Eviews, R, and S-PLUS. Class presentation on a paper topic to be assigned with a summary write-up and replication. Take home final exam.
Graduate level econometrics or equivalent (econ 580-583). Familiarity with time series methods at the level of econ 584 or stat 519. Some familiarity with statistical programming using matrix languages (Matlab, R, S-PLUS).
Taylor, S. (2005). Asset Price Dynamics, Volatility, and Prediction. Princeton University Press.
Jondeau, E., Poon, S.-H., and Rockinger, M. (2006). Financial Modeling Under Non-Gaussian Distributions, Springer-Verlag ebook. Available online from UW libraries.
Zivot, E. and Wang, J. (2006). Modeling Financial Time Series with S-PLUS, Second Edition. Springer-Verlag.
Tsay, R. (2006). Analysis of Financial Time Series, Second Edition. Wiley.
Supplemental reading from journal articles and non-required textbooks (available on course webpage)
The course will utilize Eviews, R and S-PLUS for data analysis and statistical modeling.
S-PLUS is a statistical modeling and graphical analysis program sold by Tibco Corporation (a local Seattle company). There are several add-on modules for S-PLUS. We will utilize the S+FinMetrics module for some of the statistical analysis.
The book Modeling Financial Time Series with S-PLUS by Eric Zivot and Jiahui Wang, Springer-Verlag, serves as the User's Guide for S+FinMetrics. A pdf version of the book will be given out in class.