On Fixed Width Confidence Bounds for the Difference of the Means of Two Linear Processes

  • Published : 1996.12.01

Abstract

In this article we consider a sequential procedure for the fixed width interval estimation of the means of two mutually independent linear processes. It is shown that the proposed stopping rule is asymptotically efficient as in iid samples (cf. Robbins, Simons and Starr(l967)).

Keywords

References

  1. Time Series: Theory and Mehtods, (2nd ed.) Brockwell, P. J.;Davis, R. A.
  2. Sequential Analysis v.11 Sequential estimation of the mean of a linear process Fakhre-Zakeri;Lee, S.
  3. Journfal of Multivariate Analysis v.47 Sequential estimation of the mean vector of a multivariate linear process Fakhre-Zakeri, I.;Lee, S.
  4. Stopped Random Walks: Limit Theorems and applications Gut, A.
  5. Journal of Korean Statistical Society v.23 Random central limit theorem of a stationary lattice process Lee, S.
  6. Probability and Statistics Sequential estimation of the mean of a normal population Robbins, H. E.;U. Grenandrer(ed.)
  7. Annals of Mathematical Statistics v.38 A Sequrntial analogue of the Beherens-Fisher problem Robbins, H.;Simons, G.;Starr, N.
  8. Annals of Statistics v.15 Sequential estimation of the mean of a first order stationary process Sriram, T. N.
  9. Annals of Statistics v.20 Asymptotics for linear processes Philips, P. C. B.;Solo, V.