ASYMPTOTIC MULTIVARIATE NORMALITY FOR THE SUBSERIESVALUES OF A GENERAL STATISTIC FROM A STATIONARY SEQUENCEWITH APPLICATIONS TO NONPARAMETRIC CONFIDENCE INTERVALS
E. Carlstein
Abstract: Let be a strictly stationary -mixing sequence with
unknown marginal distributions and unknown dependence structure. Suppose that, given data
the statistic is a point estimator of the
unknown parameter If a sample series is available, then the subseries values
may be used to construct a nonparametric confidence interval on
via either Student’s distribution or via the Typical Value principle. The asymptotic
justification for both methods rests upon a more general result which provides necessary and
sufficient conditions for asymptotic multivariate normality of subseries values.