A Note on Minimax Robustness of Designs Against Correlated or Heteroscedastic Responses

Douglas P. Wiens
DOI: https://doi.org/10.1093/biomet/asae001
2024-01-17
Abstract:We present a result according to which certain functions of covariance matrices are maximized at scalar multiples of the identity matrix. This is used to show that experimental designs that are optimal under an assumption of independent, homoscedastic responses can be minimax robust, in broad classes of alternate covariance structures. In particular it can justify the common practice of disregarding possible dependence, or heteroscedasticity, at the design stage of an experiment.
Statistics Theory
What problem does this paper attempt to address?