Combining Forecasts Using Constrained M-Estimators

The result of Clemen (1986) shows that the combination of unbiased forecasts by means of a LS-regression model without an intercept and with the constraint that coefficients sum to one gives less spread prediction than the general regression model. Here the result is generalized for the M -estimators of the regression model. In order to achieve this an asymptotic representation of the M -estimators is used.

MoreLess

Year of publication:	1996
Authors:	Augilar, Lucia ; Rubio, Asunción ; Víšek, Jan
Published in:	Bulletin of the Czech Econometric Society. - Česká ekonometrická společnost - CES. - Vol. 3.1996
Publisher:	Česká ekonometrická společnost - CES
Subject:	Combination of forecasts \| accuracy of prediction \| constrained M-estimators

More details

Extent:	text/html
Type of publication:	Article
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10008473453