Biases and Mean Square Errors Of Estimators Using Weibull Probability Paper

Abstract

Biases and mean square errors of estimators of parameters while using Weibull probability paper for small samples are investigated. The location and scale parameters of extreme-value distribution are estimated by least square and two point method. In the case of shape parameter of Weibull distribution only two point estimation is considered. In both the cases the problem of choosing plotting positions are studied. It has been observed that efficiency (ratio of variance of two point estimator and C. R. lower bound) for shape parameter of Weibull distribution is about 0.55-0.65 for sample size of 10-20, which is comparable with that obtainable by other methods. The method proposed in this paper can be treated as practical and efficient.