Abstract
Dunnett (1984) proposed a fixed sample-size procedure for selecting the best population among several Bernoulli experimental populations in comparison with a control Bernoulli population. Having as a reference point Dunnett's procedure (1984), we propose a closed sequential procedure that takes fewer observations but achieves the same probability of a correct selection as the original procedure.
Original language | English (US) |
---|---|
Pages (from-to) | 1085-1102 |
Number of pages | 18 |
Journal | Communications in Statistics - Theory and Methods |
Volume | 37 |
Issue number | 7 |
DOIs | |
State | Published - Jan 2008 |
Externally published | Yes |
Keywords
- Bernoulli population
- Binomial distribution
- Curtailment
- Probability of correct selection
- Ranking and selection
- Sample size
ASJC Scopus subject areas
- Statistics and Probability