A double-sequential sampling scheme

In this article, we have developed a new sequential sampling scheme, called the double-sequential sampling scheme, in which sampling is proceeded k-at-a-time (k >= 2) part of way, and then one-at-a-time sequentially. Requiring the same projected sample size as the Anscombe-Chow-Robbins (ACR) scheme, it further helps to save sampling operations. We include the statistical inference problems of fixed-width confidence interval estimation and minimum risk point estimation for a normal mean when the population variance also remains unknown as two illustrations. These are followed by extensive sets of simulations and real-data illustrations using horticulture data, showing great practical applicability of this new sequential sampling scheme.