Estimation of the parameters of the logistic distribution by sample quantiles

Abstract

Large sample estimation of the mean and the standard deviation of the logistic distribution is discussed. Best linear unbiased estimates based on k = 2(1)9 sample quantiles are used. The optimum spacings, the coefficients to be used in computing the estimates and the variances and asymptotic efficiencies of the estimates are given. A comparison is obtained between the optimum spacings and spacing the observations using the expected values of order statistics.