Two-stage procedures for comparing several exponential populations with a control when the Scale Parameters are unknown and unequal

Abstract

In this paper, two-stage procedures for comparing several exponential populations with a control are developed under the assumption of unknown and unequal variances. Conventional techniques using Bonferroni's inequality as given in Mukhopadyay and Hamdy (1984) are used to derive multiple comparison results. By exploiting recent techniques given in Lam (1987, 1988), improvements on multiple comparison results are obtained. The results can be applied to selecting the exponential population with the largest (or with the smallest) location parameter using the indifference zone approach in Bechhofer (1954) or the subset selection approach in Gupta (1956).

Keywords:

• selection and ranking
• multiple comparison
• largest location
• unequal scales
• two-stage procedures