Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/19661 
Year of Publication: 
2006
Series/Report no.: 
Discussion Paper Series 1 No. 2006,32
Publisher: 
Deutsche Bundesbank, Frankfurt a. M.
Abstract: 
This paper considers Bayesian regression with normal and doubleexponential priors as forecasting methods based on large panels of time series. We show that, empirically, these forecasts are highly correlated with principal component forecasts and that they perform equally well for a wide range of prior choices. Moreover, we study the asymptotic properties of the Bayesian regression under Gaussian prior under the assumption that data are quasi collinear to establish a criterion for setting parameters in a large cross-section.
Subjects: 
Bayesian VAR
ridge regression
Lasso regression
principal components
large cross-sections
JEL: 
C33
C13
C53
C11
Document Type: 
Working Paper

Files in This Item:
File
Size
375.39 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.