Course Director: Joann Jasiak
This graduate course in time series analysis will be offered in-person. The students are expected to have already taken courses in intermediate or advanced econometrics. The class will meet at scheduled times on Thursdays at 11:30 in Ross S 125. All course materials, including the pre-recorded lectures, will be available on E-class. The objective is to provide the students with a solid theoretical background and a selection of advanced econometric methods for later use in independent applied research. The course covers linear and nonlinear time series models with applications to macroeconomics and finance and their estimation methods. The content of the course includes: Part 1 - properties of univariate stationary processes and the Autoregressive Moving Average (ARMA) models. Part 2 - departures from stationarity, which include unit root processes and the Generalized Autoregressive Conditional Heteroskedastic (GARCH) models. Part 3 - multivariate models, such as the Vector Autoregressive (VAR) model and the Error Correction (ECM) model, causality and cointegration. The models and their applications will be illustrated by simulations and examples of time series from economics and finance. Additional examples for empirical analysis, simulations and problems will be provided to students in assignments. Suggested software are SAS, R and STATA.
Requirements, Evaluation and Other Details1. Mid-term exam in person or on-line 30% approximate date of exam: : March 03
2. Final exam in-person or on-line 50% (date to be determined)
3. Assignments 20% three sets of empirical and theoretical questions available on E-class to be handed in or e-mailed on approximately February10, March 10 and the last day of classes.
1. Introduction: time series (examples), objectives of time series analysis, model classification
2. Stochastic Processes: difference and lag operators, difference equations and their solutions, stationarity
3. Autocovariance and autocorrelation functions, Wold theorem
4. Conditional mean dynamics: ARMA models, model selection, estimation: Maximum Likelihood, and testing, forecasting, seasonality
5. Nonstationary series: deterministic and stochastic trends, unit root tests, switching regimes, spurious regressions
6. Conditional variance dynamics: GARCH models, applications, Quasi Maximum Likelihood, estimation and testing
7. Multivariate Time Series Models: VAR - estimation: Maximum Likelihood, OLS, and tests
8. Causality, exogeneity, impulse response function, variance decomposition
9. Cointegration and common trends
10. Error Correction Models (ECM) - estimation and tests
All lectures are recorded and videos are posted on e-class along with .pdf files containing the material
Books and Other Reference Materials
Enders, W., Applied Econometric Time Series 3rd or 4th ed., Wiley, 2010 or 2015 (Available on e-class for free 14 day trial)
Lecture notes at www.jjstats.com
Martin, V., Hurn, S, Harris, D., Econometric Modelling with Time Series, Cambridge Unversity Press 2013
Wei, William W.S., Time Series Analysis, Pearson, 2006 (2nd ed.).
Gourieroux, C. and A. Monfort, Time Series and Dynamic Models, Cambridge University Press, 2002.
Early Papers (easy to read) :
Bollerslev, T., R.F. Engle and D.B. Nelson (1993); "ARCH Models," in Handbook of Econometrics, Vol. 4.
Campbell, J.Y. and P. Perron, "Pitfalls and Opportunities: What Macroeconomists Should Know about Unit Roots," NBER Macroeconomics Annual, 1991, (O.T. Blanchard and S. Fisher, eds.), MIT Press.
Diebold,F.X. and M. Nerlove (1990); " Unit Roots in Economic Time Series," in Advances in Econometrics Vol 8, pp 3-69.
Nelson, C.R. and C.J. Plosser (1982), "Trends and Random Walks in Macroeconomic Time Series," Journal of Monetary Economics 10, pp. 139-162.
Sims, C.A. (1972), "Money, Income and Causality," American Economic Review 62, pp. 540-552.
Sims, C.A. (1980), "Macroeconomics and Reality," Econometrica 48, pp. 1-48.
Stock, J.H. and M.W. Watson (1988), "Testing for Common Trends," JASA 83, pp. 1097-1107.
Tiao, G.C. and G.E.P Box (1981), "Modelling Multiple Time Series with Applications," JASA 76, pp. 802-816.