Variable Selection for Large Unbalanced Datasets Using Non-Standard Optimisation of Information Criteria and Variable Reduction Methods

quantf research Working Paper Series: WP04/2014

21 Pages Posted: 2 Jun 2014

See all articles by George Kapetanios

George Kapetanios

King's College, London

Massimiliano Giuseppe Marcellino

Bocconi University - Department of Economics; Centre for Economic Policy Research (CEPR)

Fotis Papailias

Quantf Research; University of London, King's College London, Department of Management

Date Written: June 1, 2014

Abstract

We consider forecasting key macroeconomic variables using many predictors extracted from the Eurostat PEEIs dataset. To avoid the curse of dimensionality, we rely on model selection and model reduction. For model selection we use heuristic optimisation of information criteria, including simulated annealing, genetic algorithms, MC^3 and sequential testing. For model reduction we employ the methods of principal components, partial least squares and Bayesian shrinkage regression. The problem of unbalanced datasets is discussed and potential solutions are suggested. We provide adequate evidence that these methods could be useful in forecasting. Their predictive performance is evaluated in a pseudo out-of-sample exercise, comparing the results relative to a univariate AR(1) benchmark. Our findings are very encouraging for forecasting the growth rate of quarterly consumption and GDP, and monthly industrial production and inflation.

Keywords: Heuristic optimisation, Information criteria, Unbalanced datasets, Model Reduction, Forecasting, PEEI

Suggested Citation

Kapetanios, George and Marcellino, Massimiliano and Papailias, Fotis, Variable Selection for Large Unbalanced Datasets Using Non-Standard Optimisation of Information Criteria and Variable Reduction Methods (June 1, 2014). quantf research Working Paper Series: WP04/2014, Available at SSRN: https://ssrn.com/abstract=2444418 or http://dx.doi.org/10.2139/ssrn.2444418

George Kapetanios

King's College, London ( email )

30 Aldwych
London, WC2B 4BG
United Kingdom
+44 20 78484951 (Phone)

Massimiliano Marcellino

Bocconi University - Department of Economics ( email )

Via Gobbi 5
Milan, 20136
Italy

Centre for Economic Policy Research (CEPR) ( email )

London
United Kingdom

Fotis Papailias (Contact Author)

Quantf Research ( email )

London
United Kingdom

HOME PAGE: http://www.quantf.com

University of London, King's College London, Department of Management ( email )

150 Stamford Street
London, SE1 9NN
United Kingdom

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
64
Abstract Views
469
rank
418,046
PlumX Metrics