Simultaneous confidence intervals for comparing several exponential location parameters with control in direction-mixed families

AbstractIn this paper, a procedure for simultaneously comparing of k(k ≥ 2) two-parameter exponential populations with a control population with respect to location parameters in direction-mixed families of hypotheses is proposed and respective simultaneous confidence intervals are obtained. Computation of critical constants required for the implementation of the proposed procedure is discussed under two different criterion and selected values of the critical constants are tabulated. Power comparisons of the proposed procedure with the existing competitor are carried out using Monte Carlo simulation. Working of the proposed procedure and advantages are demonstrated with the help of a numerical example.