‘The use of U-statistics for testing normality against nonsymmetric alternatives’

Abstract

In §3 the values of a 1 , and a 2 and a 3 should be multiplied by ( n −1)/( n +1). Thus, the entries for T 2 n in Table 2 should be multiplied by {( n )−1)/( n +1)} 1.2 . With this correction, it appears that T 2 n approaches normality at about the same rate as T 1 n , perhaps faster. A revised version of Table 1 for T 2 n follows.