Model selection for panel data forecasting

Rt Greenaway-McGrevy

No 38620, Working Papers from Department of Economics, The University of Auckland

Abstract: This paper develops new model selection methods for forecasting panel data using a set of least squares (LS) vector autoregressions. Model selection is based on minimizing the estimated quadratic forecast risk among candidate models. We provide conditions under which the selec- tion criterion is asymptotically e¢ cient in the sense of Shibata (1980, Ann. Statist. 8, 147-164) as n (cross sections) and T (time series) approach in nity. Relative to extant selection criteria, this criterion places a heavier penalty on model dimensionality in order to account for the e¤ects of parameterized forms of cross sectional heterogeneity (such as xed e¤ects) on forecast loss. We also extend the analysis to bias-corrected least squares, showing that signi cant reductions in forecast risk can be achieved.

Date: 2017
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/2292/38620

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:auc:wpaper:38620

Access Statistics for this paper

More papers in Working Papers from Department of Economics, The University of Auckland Contact information at EDIRC.
Bibliographic data for series maintained by Library Digital Development ().