Cover Image

Log versus level in VAR forecasting: 16 Million empirical answers - expect the unexpected

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 setup and approximately 16 Mio. pseudo-out-of-sample forecasts are evaluated. The results show that, on average, the knee-jerk transformation of the data is at best harmless.
Year of publication:
2007
Authors: Mayr, Johannes ; Ulbricht, Dirk
Publisher: Munich : ifo Institute - Leibniz Institute for Economic Research at the University of Munich
Subject: VAR-forecasting | logarithmic transformation
Series:
Type of publication: Book / Working Paper
Type of publication (narrower categories): Working Paper
Language: English
Other identifiers:
648826732 [GVK]
hdl:10419/73744 [Handle]
RePec:ces:ifowps:_42 [RePEc]
Classification: C52 - Model Evaluation and Testing ; C53 - Forecasting and Other Model Applications
Source:
Persistent link: https://ebvufind01.dmz1.zbw.eu/10010312097
Similar items by subject
Similar items by person