Interval Estimation of the Overlapping Coefficient of Two Normal Distributions: One Way ANOVA with Random Effects

We consider the problem of deriving confidence interval for the Overlapping Coefficient (OVL) between one way random model with balanced and unbalanced data. The confidence intervals are derived using the concept of a Generalized Pivotal Quantity (GPQ). The accuracy of the proposed solutions are assessed using estimated coverage probabilities, and are also compared with other approximate solutions. The numerical results show that the GPQ method performs well in the estimation of OVL of two normal distributions under one way random model set-up. The results are illustrated with simulated examples.