Empirical bayes estimation of the mean of a “nonhomogeneous” finite population
Ram Chandra Tiwari, Parthasarathi Lahiri · Communication in Statistics- Theory and Methods · 1989
Suppose , where Yidenotes the value of a characteristic of interest associated with the unit i(i=1,..., N). Ghosh and Meeden (1986) considered an empirical Bayes approach to the estimation of the finite population mean, under squared error loss and the model i=∊+∊i(i=1,…N), where θ, ∊1,…,∊n are independently distributed withand . In this paper we extend their results to the case where ∊i's are independent with , and , where 0 < c ≤ 1.