Template-type: ReDIF-Paper 1.0
Author-Name: Ralf Brüggemann
Author-Email: brueggem@wiwi.hu-berline.de
Author-Workplace-Name: Humboldt-Universität zu Berlin, Germany
Author-Name: Hans-Martin Krolzig
Author-Email: hans-martin.krolzig@nuf.ox.ac.uk
Author-Workplace-Name: Department of Economics, and Nuffield College, Oxford University
Author-Name: Helmut Lütkepohl
Author-Email: helmut.luetkepohl@iue.it
Author-Workplace-Name: Humboldt-Universität zu Berlin and European University Institute, Italy
Title: Comparison of Model Reduction Methods for VAR Processes
Abstract:
  The objective of this study is to compare alternative
  computerized model-selection strategies in the context of the
  vector autoregressive (VAR) modeling framework. The focus is on a
  comparison of subset modeling strategies with the
  general-to-specific reduction approach automated by PcGets.
  Different measures of the possible gains of model selection are
  considered: (i) the chances of finding the `correct' model, that
  is, a model which contains all necessary right-hand side variables
  and is as parsimonious as possible, (ii) the accuracy of the
  implied  impulse-responses and (iii) the forecast performance of
  the models obtained with different specification algorithms. In
  the Monte Carlo experiments, the procedures recover the DGP
  specification from a large VAR with anticipated size and power
  close to commencing from the DGP itself when evaluated at the
  empirical size. We find that subset strategies and PcGets are
  close competitors in many respects, with the forecast comparison
  indicating a clear advantage of the PcGets algorithm.
Classification-JEL: C32, C51
Keywords: Model selection, Vector autoregression, Subset model, 
  Lag order determination, Data mining
Length:20 pages
Creation-Date: 2003-04-02
Number:2003-W13
File-URL: http://www.nuff.ox.ac.uk/economics/papers/2003/W13/bkl02.pdf
File-Format: application/pdf
Handle: RePEc:nuf:econwp:0313