A confidence distribution approach to inferring the among-group variance component in one-way random effects model with unequal error variances

Abstract

In this paper, we construct a new kind of confidence intervals for the among-group variance component through constructing confidence distribution in one-way random effects model with unequal error variances. At first, a combined asymptotic confidence distribution is constructed for the parameter of interest, then the confidence interval is derived naturally by the property of confidence distribution. Simulation results demonstrate that the new confidence intervals perform very well in terms of both empirical coverage probability and average interval length. Finally, a real data example is illustrated.