Theoretical Mathematics & Applications

Gram Charlier and Edgeworth expansion for sample variance

  • Pdf Icon [ Download ]
  • Times downloaded: 10500
  • Abstract

    In this paper, we derive a valid Edgeworth expansions for the Bessel corrected empirical variance when data are generated by a strongly mixing process whose distribution can be arbitrarily. The constraint of strongly mixing process makes the problem not easy. Indeed, even for a strongly mixing normal process, the distribution is unknown. Here, we do not assume any other assumption than a sufficiently fast decrease of the underlying distribution to make the Edgeworth expansion convergent.
    This results can obviously apply to strongly mixing normal process and provide an alternative to the work of [6] and [4].

    Mathematics Subject Classification: 62E10; 62E15
    Keywords: sample variance; Edgeworth expansion