If , asymptotic normality gives an asymptotically level- test that rejects whenFor the median, . For Huber,For the trimmed mean, with ,These tests have bounded influence functions, so a small contamination proportion has bounded first-order effect on their statistics, asymptotic levels, and powers. Their finite-contamination protection is quantified by the breakdown points in part (c).
Articles by others on the same topic
There are currently no matching articles.