Edgeworth expansion for Student's t statistic under minimal moment conditions (Q580839): Difference between revisions

Let T be the t-statistic based on a random sample \(X_ 1,X_ 2,...,X_ n\), and denote by \(\Phi\) (x) and \(\phi\) (x) the distribution and density functions, respectively, of the standard normal distribution. It follows from \textit{R. N. Bhattacharya} and \textit{J. K. Ghosh} [Ann. Stat. 6, 434- 451 (1978; Zbl 0396.62010)] that, if \(E(X_ 1^{2k+2})<\infty\) for some \(k\geq 1\), then the Edgeworth expansion \[ P(T\leq y]=\Phi (y)+\sum^{k}_{i=1}n^{-i/2}P_ i(y)\phi (y)+o(n^{-k/2}) \] obtains. Hence \(P_ i(y)\) is a polynomial of degree 3i-1 whose coefficients depend only on \(\mu_ 1,...,\mu_{i+2}\). It is now shown that the above expansion is true when merely \(E(X_ 1^{k+2})\) is finite if the distribution of \(X_ 1\) is non-singular.