Log versus Level in VAR Forecasting: 42 Million Empirical Answers - Expect the Unexpected

16 Pages Posted: 21 Oct 2014

See all articles by Johannes Mayr

Johannes Mayr

Bayerische Landesbank

Dirk Ulbricht

German Institute for Economic Research (DIW)

Date Written: October 2014

Abstract

The use of log-transformed data has become standard in macroeconomic forecasting with VAR models. However, its appropriateness in the context of out-of-sample forecasts has not yet been exposed to a thorough empirical investigation. With the aim of filling this void, a broad sample of VAR models is employed in a multi-country set up and approximately 42 Mio. pseudo-out-of-sample forecasts of GDP are evaluated. The results show that, on average, the knee-jerk transformation of the data is at best harmless.

Keywords: VAR-forecasting, Logarithmic transformation

JEL Classification: C52, C53

Suggested Citation

Mayr, Johannes and Ulbricht, Dirk, Log versus Level in VAR Forecasting: 42 Million Empirical Answers - Expect the Unexpected (October 2014). DIW Berlin Discussion Paper No. 1412. Available at SSRN: https://ssrn.com/abstract=2512166 or http://dx.doi.org/10.2139/ssrn.2512166

Johannes Mayr

Bayerische Landesbank ( email )

Brienner Strasse 18
80333 Munich
United States

Dirk Ulbricht (Contact Author)

German Institute for Economic Research (DIW) ( email )

Mohrenstra├če 58
Berlin, 10117
Germany

Register to save articles to
your library

Register

Paper statistics

Downloads
18
Abstract Views
356
PlumX Metrics