Robust designs for diallel crosses against the missing of one block

Abstract

Dey & Midha (1996) showed that some of the complete diallel crosses plans, obtained by using triangular partially balanced designs with two associate classes, are optimal. In this investigation, it is derived that these optimal designs for diallel crosses are robust also against the unavailability of one block.

Robustness of a Class of Partial Diallel Cross Designs to the Unavailability of a Complete Block of Observations

“…The Effect of a Missing Block in a CDC Design Ghosh & Desai (1998, 1999 and Ghosh & Biswas (2000) have examined the effects of losing a complete block of observations in CDC designs derived in the manner described earlier. Before we proceed to investigate the robustness of PDC designs to the loss of a block of crosses, we consider examples of CDC designs to illustrate the way in which, even in the case of the complete diallel cross designs, different efficiency-losses occur for different line comparisons when a block of the design becomes unavailable.…”

“…They concluded, on the basis of A-efficiency, that such designs are fairly robust to missing data. The robustness of optimal CDC designs to the loss of observations from a complete block was considered by Ghosh & Biswas (2000) with similar 146 R. Mansson & P. Prescott conclusions. They showed, for a variety of designs subject to a missing block, that the A-efficiencies ranged from 91% to 98%.…”