‘Some incomplete block designs of maximum efficiency’

Abstract

We are most grateful to Professor J. Kiefer for drawing our attention to a paper by Takeuchi (1961) in which optimality is considered, in terms of supremum of variance of estimates for treatment contrasts and in terms of maximum variance of estimates of treatment effect differences. These criteria lead to the same designs as did our criterion of optimizing the average variance of estimates of treatment effect differences. Our result can alternatively be obtained by a development from Takeuchi's paper. TAKEUCHI, K. (1961). Rep. Statist. Appl. Res. J.U.S.E. 8, 28–33.