A BAYESIAN APPROACH IN THE DETECTION OF OUTLIERS FOR NORMAL AND WEIBULL DISTRIBUTIONS

Jorge Alberto ACHCAR[1]

§    ABSTRACT: The problem of detection of outliers is of great practical interest. In this paper, we present some Bayesian methods for detection of outliers for normal and Weibull distributions combining usual graphical procedures with posterior densities for prediction values and predictive densities for discordant observations.

§    KEYWORDS: Outliers; normal distribution; Weibull distribution; posterior densities; predictive densities.

 



[1] ICMSC-USP - 13560-970 - São Carlos - SP.