Multivariate Time Series Analysis and Forecasting

I. Syllabus:

The first part of this course offers a detailed introduction to the vector autoregressive (VAR) model, the workhorse model for multivariate time series analysis. This includes issues of specification, estimation, forecasting, and (structural) interpretation. The second part of this course looks more deeply into important topics of forecasting such as loss functions, finding optimal forecasts, evaluating forecasts, and forecast combinations.

  • Stable Vector Autoregressive Processes
  • Estimation of Vector Autoregressive Processes
  • VAR Order Selection and Checking the Model Adequacy
  • Topics in Forecasting


II. Prerequisites:


III. Exam:

IV. Downloads:


V. Materials:

  • Textbook: Helmut Lütkepohl (2007), New Introduction to Multiple Time Series Analysis, Springer-Verlag, Berlin.(Main reference)
  • F.X. Diebold and R.S. Mariano (1995) Comparing predictive accuracy, Journal of Business and Economic Statistics 13, 253-263.
  • G. Elliott and A. Timmermann (2008) Economic forecasting, Journal of Economic Literature 46, 3-56.
  • A. Timmermann (2006) Forecast combinations, in: Handbook of Economic Forecasting, Vol. 1, 99-134.
  • More papers will be announced in the lecture.


VI. Lecture:

VII. Voluntary tutorial (PC-LAB):

Access to the computer lab requires one-time registration with a Stu-Account

VIII. Tutorial (paper/pen):