Citation

Discussion Paper Details

Please find the details for DP7139 in an easy to copy and paste format below:

Full Details   |   Bibliographic Reference

Full Details

Title: Variable Selection and Inference for Multi-period Forecasting Problems

Author(s): M Hashem Pesaran, Andreas Pick and Allan Timmermann

Publication Date: January 2009

Keyword(s): factor-augmented VAR, forecast horizon and macroeconomic forecasting

Programme Area(s): Financial Economics and International Macroeconomics

Abstract: This paper conducts a broad-based comparison of iterated and direct multi-step forecasting approaches applied to both univariate and multivariate models. Theoretical results and Monte Carlo simulations suggest that iterated forecasts dominate direct forecasts when estimation error is a first-order concern, i.e. in small samples and for long forecast horizons. Conversely, direct forecasts may dominate in the presence of dynamic model misspecification. Empirical analysis of the set of 170 variables studied by Marcellino, Stock and Watson (2006) shows that multivariate information, introduced through a parsimonious factor-augmented vector autoregression approach, improves forecasting performance for many variables, particularly at short horizons.

For full details and related downloads, please visit: https://cepr.org/active/publications/discussion_papers/dp.php?dpno=7139

Bibliographic Reference

Pesaran, M, Pick, A and Timmermann, A. 2009. 'Variable Selection and Inference for Multi-period Forecasting Problems'. London, Centre for Economic Policy Research. https://cepr.org/active/publications/discussion_papers/dp.php?dpno=7139