Estimators Based on Order Statistics of Small Samples from a Three-Parameter Lognormal Distribution

Abstract

The expectations of order statistics of a logarithmic-normal distribution are expressed as functions which are linear in a location and a scale parameter but nonlinear in a shape parameter of the distribution. The regression of the observed small-sample order statistics on approximations to their expectations is considered and an iterative procedure is proposed which leads to approximate weighted least squares estimates of all three parameters.The performance of the iterative procedure as well a3 the properties of the estimators are studied by means of Monte Carlo simulation experiments involving samples of size 10, 15 and 20 from standardized distributions (zero median, unit geometric mean) with shape parameter (natural logarithmic variance) in the range zero to unity.It is shown that the iterative procedure always converges and that almost unbiased estimates may be obtained. The variances of the location and scale parameter estimates compare favorably with those of the well-known order-statistic estimates of a standard normal distribution. The parameterization of the lognormal distribution which is adopted leads naturally to use of the latter estimates in the limiting case of samples which exhibit symmetry or even negative skewness.While only complete samples are considered explicitly, the simple extension required to take care of censored samples is indicated.

Towards a low CO2 emission building material employing bacterial metabolism (2/2): Prospects for global warming potential reduction in the concrete industry

“…Inductive inference is inherently uncertain, as discussed by Gilboa [36]. We estimate the quantitative input data (see also section 2.7) based on the Dempster-Shafer theory [37] and fit the data to the three-parameter lognormal distribution function proposed by Munro and Wixley (1970) [38]: …”

On the consistency of the maximum likelihood estimator for the three parameter lognormal distribution

“…This model has been widely used in applications and its estimation approaches have been studied by many, including Cohen (1951), Hill (1963), Harter and Moore (1966), Munro and Wixley (1970), Giesbrecht and Kempthorne (1976), Cohen and Whitten (1980), Crow and Shimizu (1998) and Basak et al (2009), among others. But the theoretical properties of the proposed methods were not fully addressed rigorously in these papers.…”

Easy estimation by a new parameterization for the three-parameter lognormal distribution

“…For problems to seek α, that is, the one-parameter estimation problems for the lognormal distribution, Wingo (1975Wingo ( , 1976Wingo ( , 1984 has proposed a computing method to avoid the singular range x n − α ≤ δ by adopting a penalty function. On the other hand, for problems to seek simultaneously α, β and γ, where ( α, β, γ) is a PRM, that is, the three-parameter estimation problems (Lambert, 1964) for the distribution, Munro and Wixley (1970) have proposed a parameterization to improve the convergency of many iterative methods (Eastham, LaRiccia and Schuenemeyer, 1987;Hirose, 1997). As seen now, there are two ways for dealing with the parameter estimation of the distribution for complete data.…”