Template-Type: ReDIF-Paper 1.0
Author-Name: Heij, C.
Author-Name-Last: Heij
Author-Name-First: Christiaan
Author-Name: Groenen, P.J.F.
Author-Name-Last: Groenen
Author-Name-First: Patrick
Author-Person: pgr229
Author-Name: van Dijk, D.J.C.
Author-Name-Last: van Dijk
Author-Name-First: Dick
Author-Person: pva27
Title: Forecast comparison of principal component regression and principal covariate regression
Abstract: Forecasting with many predictors is of interest, for instance, in macroeconomics and finance. This paper compares two methods for dealing with many predictors, that is, principal component regression (PCR) and principal covariate regression (PCovR). The forecast performance of these methods is compared by simulating data from factor models and from regression models. The simulations show that, in general, PCR performs better for the first type of data and PCovR performs better for the second type of data. The simulations also clarify the effect of the choice of the PCovR weight on the orecast quality.
Creation-Date: 2005-08-02
File-URL: https://repub.eur.nl/pub/6918/ei200528.pdf
File-Format: application/pdf
Series: RePEc:ems:eureir
Number: EI 2005-28
Keywords: economic forecasting, factor model, principal components, principal covariates, regression model
Handle: RePEc:ems:eureir:6918