a Department of Mathematics and Statistics, McMaster University, 1280 Main Street West, Hamilton, Ont., Canada L8S 4K1
b Department of Statistics, Loyola College, Chennai 600 034, India
Abstract:
Thomas and Wilson (Technometrics 14 (1972) 679) developed a computational method for calculating the single and product moments of order statistics from progressively censored samples by making use of the corresponding moments of the usual order statistics. The absence of an explicit representation for the marginal and joint density function of order statistics under progressive censoring makes their method extremely tedious. By deriving the required marginal and joint density functions in explicit form, we obtain an alternative, highly efficient, method for computing the desired moments.