Multi-step forecasting with large vector autoregressions

Andreas Pick, Matthijs Carpay

Research output: Contribution to journalArticleAcademicpeer-review

63 Downloads (Pure)

Abstract

This chapter investigates the performance of different dimension reduction approaches for large vector autoregressions in multi-step ahead forecasts. The authors consider factor augmented VAR models using principal components and partial least squares, random subset regression, random projection, random compression, and estimation via LASSO and Bayesian VAR. The authors compare the accuracy of iterated and direct multi-step point and density forecasts. The comparison is based on macroeconomic and financial variables from the FRED-MD data base. Our findings suggest that random subspace methods and LASSO estimation deliver the most precise forecasts.
Original languageEnglish
Pages (from-to)73-98
JournalAdvances in Econometrics
Volume43A
DOIs
Publication statusPublished - 18 Jan 2022

Fingerprint

Dive into the research topics of 'Multi-step forecasting with large vector autoregressions'. Together they form a unique fingerprint.

Cite this