Time Series Econometrics

Professor F.X. Diebold

Spring 2016

This course provides a Ph.D.-level introduction to econometric time-series analysis.

Text: Diebold's *Time Series Econometrics.* Book, slides, code, etc. here.

A few papers to read here. Useful additional books here. Useful software here.

Format: Lectures that stress applied econometric theory.

Topics:

The Wold Representation and its Approximation

Markovian Structure, State Space, and the Kalman Filter

Likelihood Evaluation and Optimization

Simulation Methods and Their Application

Bayesian Estimation via Markov Chain Monte Carlo

Integration, Cointegration and Long Memory

Nonlinear/Non-Gaussian State Space

Conditional Variance Dynamics

Big Data and High Dimensionality

Office hours (held in McNeil 519) are listed at http://www.ssc.upenn.edu/~fdiebold.

Teaching assistants will be heavily involved, including small-scale help by email and large-scale help in weekly review/supplementary sessions. (Office hours and review session times and locations, contact info, etc., to be announced.)

Grading: *N* problem sets (each 60/*N* %) and a final exam ("practice prelim") (40%). Good performance is crucially dependent on regular class preparation, attendance and participation.

Problem Set 1 (Due February 25.) Use the monthly U.S. housing starts and completions data supplied by the TA's, reserving the last six observations for out-of-sample forecast comparisons. Discuss all results as appropriate. First graph the data. Then do the following. (1) (Univariate) Fit an AR(4) to completions, use it for honest Wiener-Kolmogorov forecasting of the six hold-out completions observations, and assess accuracy. (2) (Multivariate) Estimate and interpret the autocorrelation function; Model the two series jointly as a VAR(4); Calculate the autocorrelation function implied by your estimated VAR and compare it to the one directly estimated earlier; Perform a Granger-causality analysis; Using Cholesky factor identification, calculate and graph the full set of impulse-response functions; Again forecast the six hold-out completions observations, assess accuracy, and compare the multivariate accuracy to your earlier-assessed univariate accuracy.

Problem Set 2 (Due April 26.) Problem set is here. Also see the FRB Philadelphia web site.

Final exam date: Monday, May 9, 12-2, MCNB 286-7 (standard university date/time/location).

Note well that modifications and adjustments to this syllabus / web site are inevitable and may be implemented at any time. Check frequently for updates.