000			01759nam a22003257a 4500
003			arcduce
005			20220331220118.0
007			ta
008			131122s2006 gw_||||| |||| 00| 0 eng d
020			_a3865582079
040			_aarcduce _carcduce
082	0		_221 _a330.015195
100	1		_95522 _aDe Mol, Christine
245	1	0	_aForecasting using a large number of predictors : _bis bayesian regression a valid alternative to principal components? / _cChristine De Mol, Domenico Giannone, Lucrezia Reichlin.
260			_aFrankfurt am Main : _bDeutsche Bundesbank, _c2006
300			_a36 p.
490	1		_aDiscussion paper. Series 1: Economic studies ; _vno. 32/2006
504			_aBibliografía: p. 17-19.
520	3		_aThis 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.
650	1	4	_aANALISIS BAYESIANO _95402
653		4	_aPRONOSTICOS ECONOMICOS
653		4	_aPREDICCIONES ECONOMICAS
650		4	_92897 _aPREVISIONES ECONOMICAS
700	1		_95523 _aGiannone, Domenico
700	1		_95524 _aReichlin, Lucrezia, _d1954-
830		0	_94690 _aDiscussion paper (Deutsche Bundesbank). _pSeries 1: economic studies ; _vno. 32/2006
856	4		_uhttp://econstor.eu/bitstream/10419/19661/1/200632dkp.pdf
942			_2ddc _cDOCU _jF 330.015195 D 21081
945			_aBEA _c2014-04-07
999			_c23954 _d23954