Uniform inference in vector autoregressive models

Seminar abstract

This paper establishes uniform inference results in vector autoregressive (VAR) models. We extend Mikusheva's (2007, 2012) results for univariate autoregressive models to VAR models. We consider confidence intervals for structural impulse responses based on the Wald statistic of Lutkepohl, Staszewska-Bystrova, and Winker (2015). Our results are uniform over the parameter space and horizons. In Monte Carlo simulation experiments, we compare our method with Montiel-Olea and Plagborg-Moller's method based on local projections. Our Monte Carlo results show that our confidence intervals tend to be more accurate than theirs in terms of coverage rates.

How to attend this seminar

This seminar is free to attend.

We ask you to place contact Professor Rob Taylor for the passcode to gain access to this seminar online.

Speaker bio

Atsushi Inoue is currently Cornelius Vanderbilt Chair and Professor of Economics at Vanderbilt University and is an econometrician. He is interested in developing econometric methods for macroeconomic models, such as structural VAR models and DSGE models. Originally from Japan, he received a Ph.D. in economics from the University of Pennsylvania in 1998. He previously taught at North Carolina State University, Southern Methodist University, and the University of British Columbia before joining Vanderbilt in 2014. He currently serves as co-editor for the Journal of Business and Economic Statistics.