LINEAR ALLOCATION MODELS FOR SYMMETRIC DISTRIBUTION IN RANKED SET SAMPLING

Ranked Set Sampling (RSS) is one method to potentially increase precision and reducecost by using simple judgment or qualitative information. For symmetric distributions, an optimalallocation model was suggested by Kaur et al. (1995) (for simplicity in notation we call it by KPTmodel). This allocation model measures either only mid or extreme rank orders. This results in anestimator, which is not sufficient and hence unreliable in most of the situations, although it ismore precise then Neyman’s allocation.In this paper, we have proposed a Linear allocation model for two classes of symmetricdistributions. These two classes of symmetric distribution are mound shaped and U-shaped,depending upon the plots of the variances of the order statistics against the rank order. Theproposed allocation model is opposite to the Neyman allocation model and has an advantage overKPT model in the sense that measurements are made upon each rank orders.