Please use this identifier to cite or link to this item:
Dette, Holger
Melas, Viatcheslav B.
Year of Publication: 
Series/Report no.: 
Technical Report 2008,19
If a model is fitted to empirical data, bias can arise from terms which are not incorporated in the model assumptions. As a consequence the commonly used optimality criteria based on the generalized variance of the estimate of the model parameters may not lead to efficient designs for the statistical analysis. In this note some general aspects of all-bias designs are presented, which were introduced in this context by Box and Draper (1959). We establish sufficient conditions such that a given design is an all-bias design and illustrate these in the special case of spline regression models. In particular our results generalize recent findings of Woods and Lewis (2006).
all-bias design
spline regression models
robust designs
quadrature formulas
Document Type: 
Working Paper

Files in This Item:
167.33 kB

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