Subscribe to thousands of academic journals for just $40/month
Read and share the articles you need for your research, all in one place.

Optimal designs in regression problems with a general convex loss function

Biometrika , Volume 55 (1): 53 – Mar 1, 1968


Oxford University Press
Copyright © 1968 Oxford University Press
Publisher site
See Article on Publisher Site

Preview Only

Expand Tray Hide Tray

Optimal designs in regression problems with a general convex loss function


Abstract SUMMARY The polynomial regression model in which the errors are independent and identically distributed, is considered, and attention is focused on the estimate of β 8 , the coefficient, of the term of highest degree. A design D is a set of n values in −1, 1 of the concomitant, variable x , and the least-squares estimator of β 8 , when the design D is used, is donoted by β( D ). Then D * is said to dominate D if, for every continuous convex function c , Ec{β(D * )})≤= Ec{β(D)}. Does there exist a design D which dominates all others in this sense ? In general the answer is no, though for certain values of n and any symmetric error distribution the answer is yes. A crucial property of Tchebycheff designs emerges from discussion of this problem and this supports their use in practice. In addition two useful methods for comparing different designs are suggested. © Oxford University Press
Loading next page...

Preview Only. This article cannot be rented because we do not currently have permission from the publisher.