The problem of selecting the normal population having the largest mean has been widely studied in the literature by treating the population means and the variances as unrelated parameters. But it is very common in practice to find that the population standard deviations are related to the means by a proportionality relation; the constant of proportionality being known as the coefficient of variation. In this paper single-stage ranking and selection procedures are proposed for the above situation where the populations under study have a common known coefficient of variation. The indifference-zone approach and the subset selection approach are both considered. The large sample theory is studied in detail and the corresponding tables are provided for implementing the proposed procedures. The small sample theory is discussed briefly in the Appendix.
|Journal||Sankhyā: The Indian Journal of Statistics, Series B|
|State||Published - Jun 1978|