A Note on Favorable Estimation when Data is Contaminated
Abstract
The aim of this paper is to describe an alternative method of estimation of severities in risk assessment through the t-score moment method when the data are contaminated. The method is a bit less efficient than the standard maximum likelihood method, however it is more robust. Simulation study emphasize the fact that crucial role is played by level of contamination. In cases of very small contamination, both methods give similar results. The larger contamination, the better are the t-moment estimates. The reason for it is the bounded t-score of heavy-tailed distributions (and, consequently, bounded influence functions of the estimators).