Robust Multivariate Tolerance Regions: Influence Function and Monte Carlo Study

BOENTE, GRACIELA; FARALL, ANDRÉS

TECHNOMETRICS

The American Society for Quality and The American Statistical Association

Lugar: Washington; Año: 2008 vol. 50 p. 487 - 500

0040-1706

In this paper, we define a class of multivariate tolerance regions which turn out to be more resistant than the classical ones to outliers. The tolerance factors are numerically evaluated under the central model and the sensitivity to deviations from the normal distribution for moderate samples is studied through a Monte Carlo study. Besides, the influence function of the coverage probability allows us to compare the sensitivity of different proposals to anomalous data. Finally, real data examples are discussed.