Minimum K-S estimator using PH-transform technique

Abstract

In this paper, we propose an improvement of the Minimum Kolmogorov-Smirnov (K-S) estimator using proportional hazards transform (PH-transform) technique. The data of experiment is 47 fire accidents data of an insurance company in Thailand. This experiment has two operations, the first operation, we minimize K-S statistic value using grid search technique for nine distributions; Rayleigh distribution, gamma distribution, Pareto distribution, log-logistic distribution, logistic distribution, normal distribution, Weibull distribution, lognormal distribution, and exponential distribution and the second operation, we improve K-S statistic using PHtransform. The result appears that PH-transform technique can improve the Minimum K-S estimator. The algorithms give better the Minimum K-S estimator for seven distributions; Rayleigh distribution, logistic distribution, gamma distribution, Pareto distribution, log-logistic distribution, normal distribution, Weibull distribution, log-normal distribution, and exponential distribution while the Minimum K-S estimators of normal distribution and logistic distribution are unchanged

Keywords

• Minimum K-S estimator
• PH-transform technique