Forecasting using a large number of predictors : is bayesian regression a valid alternative to principal components? / Christine De Mol, Domenico Giannone, Lucrezia Reichlin.
Tipo de material: TextoSeries Discussion paper (Deutsche Bundesbank). Series 1: economic studies ; ; no. 32/2006Detalles de publicación: Frankfurt am Main : Deutsche Bundesbank, 2006Descripción: 36 pISBN:- 3865582079
- 21 330.015195
Tipo de ítem | Biblioteca actual | Signatura topográfica | Estado | Fecha de vencimiento | Código de barras | |
---|---|---|---|---|---|---|
Documento | Biblioteca Manuel Belgrano | F 330.015195 D 21081 (Navegar estantería(Abre debajo)) | Disponible | 21081 F |
Bibliografía: p. 17-19.
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.
No hay comentarios en este titulo.