In this paper we discuss two sequential procedures for constructing confidence intervals for the mean with a relative width requirement. Carefully stating the procedure proposed by Nadas and the procedure considered by Thomas, Iglehart, Robinson, Lavenberg and Sauer, and Law, we give some efficiency and consistency results concerning the latter, and compare their small-sample performances on a variety of stochastic models for which analytic results are available.
No takes yet. Share an insight, caveat, or question.
Law et al. (1981) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: