Estimation for the half-logistic distribution based on generalized progressive hybrid censoring
Sung-Ok Lee 1 · Suk-Bok Kang 2
12 Department of Statistics, Yeungnam University
Received 8 May 2018, revised 8 June 2018, accepted 11 June 2018
Abstract
We obtain maximum likelihood estimator (MLE) and approximate maximum like- lihood estimators (AMLEs) of the scale parameter in half-logistic distribution under generalized progressive hybrid censored samples. We also obtain a Bayes estimator of the scale parameter under squared error loss function. Finally, we examine the validity of the proposed estimators using simulated data and real data. AMLEs are obtained explicitly with closed form and AMLEs are more efficient than MLE. Bayes estimator of the scale parameter is more efficient than MLE and AMLEs.
Keywords: Approximate maximum likelihood estimators, generalized progressive hybrid censored sample, half-logistic distribution, maximum likelihood estimator.
1. Introduction
The probability density function f (x) and the cumulative distribution function F (x) of a random variable X having a half-logistic distribution with the scale parameter σ are given by;
f (x; σ) = 2 exp − σ x
σ 1 + exp − x σ 2 , x ≥ 0, σ > 0 (1.1) and
F (x; σ) = 1 − exp − x σ
1 + exp − x σ , x ≥ 0, σ > 0. (1.2) Several authors have studied about inference for the half-logistic distribution. Balakrishnan (1985) founded a few relapse relations with regard to the moments of the half-logistic dis- tribution. He also founded product moments of order statistics. Kang et al. (2008) obtained MLE and AMLEs of the scale parameter in the half-logistic distribution under progressive
1
Graduate student, Department of Statistics, Yeungnam University, Gyeongsan 38541, Korea.
2