# zbMATH — the first resource for mathematics

Truncation of the Bechhofer-Kiefer-Sobel sequential procedure for selecting the multinomial event which has the largest probability. II: Extended tables and an improved procedure. (English) Zbl 0601.62034
[For part I see ibid. 14, 283-315 (1985; Zbl 0583.62023).]
This paper is concerned with the truncated version of a vector-at-a-time sequential sampling procedure proposed by the first author, J. Kiefer and M. Sobel [Sequential identification and ranking procedures. (1968; Zbl 0208.446)] for selection of the multinomial event which has the largest probability. A greatly expanded set of truncation numbers $$(n_ 0)$$ necessary to implement the truncated (closed) procedure and the associated performance characteristics are presented.
An improvement of the procedure is also presented. It is shown that the distribution of n for the improved procedure is stochastically smaller than for the original truncated procedure for almost all values of the specified constants.
Reviewer: K.Uosaki

##### MSC:
 62F07 Statistical ranking and selection procedures 62L10 Sequential statistical analysis
Full Text:
##### References:
 [1] DOI: 10.1080/03610918508812440 · doi:10.1080/03610918508812440 [2] DOI: 10.1080/03610918508812441 · Zbl 0583.62023 · doi:10.1080/03610918508812441 [3] Bechhofer R.E., Sequential Identification and Ranking Procedures (with special reference to Koopman-Darmois populations (1968) · Zbl 0208.44601 [4] Bechhofer R.E., Proceedings of the Third Purdue Symposium on Statistical Decision Theory and Related Topics 1 pp 61– (1982) [5] DOI: 10.1080/03610928308828647 · Zbl 0537.62019 · doi:10.1080/03610928308828647 [6] DOI: 10.1093/biomet/66.1.171 · doi:10.1093/biomet/66.1.171
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.