Removal of the points that do not support an E-optimal experimental design

Publication date: Available online 14 December 2018Source: Statistics & Probability LettersAuthor(s): Radoslav Harman, Samuel RosaAbstractWe propose a method for removing design points that cannot support any E-optimal experimental design of a linear regression model with uncorrelated observations. The proposed method can be used to reduce the size of some large E-optimal design problems such that they can be efficiently solved by semidefinite programming. This paper complements the results of Pronzato [Pronzato, L., 2013. A delimitation of the support of optimal designs for Kiefer’s ϕp-class of criteria. Statistics & Probability Letters 83, 2721–2728], who studied the same problem for analytically simpler criteria of design optimality.

Source: Statistics and Probability Letters - December 15, 2018 Category: Statistics Source Type: research