摘要:

The problem of selecting the component having the largest population mean in a k-variate normal population is discussed when all the means, variances and correlations are unknown. The indifference-zone formulation has been adopted, and we propose a two-stage procedure (PS) that guarantees the nominal value (P*) of probability of correct selection under the assumption that all the correlations are non-negative. The procedure PSis the only method thus far proposed that will accomplish the desired objective. It is also of some interest that PSis asymptotically (as δ*↛ 0) "better" than the procedure PRof Dudewicz and Dalal (1975) for k = 2, P*≤.95, while for k≥3 the procedure PRis "better" than PSwhen ρij= 0, i≠j. This comparison is naturally valid when both PRand PSare applicable, and of course there are situations where only PSis applicable.

ISSN:0747-4946    

DOI:10.1080/07474948408836049

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor

摘要:

Leti=1,2,...,where the x's are non-stochastic and e1,e2,... are iid N(0, σ2) random variables, and α,β and σ are unknown parameters. Confidence sequences for α and β are constructed with the aid of invariance and a number of interesting transformations. The ensuing sequential tests with power one are also briefly discussed.

ISSN:0747-4946    

DOI:10.1080/07474948408836050

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor

摘要:

Bechhofer and Kulkarni (1982a) proposed a sequential procedure for selecting the best of k≥2 Bernoulli populations, and in a subsequent paper (1982b) gave an upper bound for the expected number of observations taken from each population by this procedure. In this note we present an asymptotically correct approximation to the expected sample size taken from each population and a slightly improved upper bound on these expected sample sizes.

ISSN:0747-4946    

DOI:10.1080/07474948408836051

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor

摘要:

In this paper, we discuss two-stage procedures for selecting the largest location parameter among k (≥ 2) negative exponential populations having unknown and unequal scale parameters. We compare this new two-stage procedure for k = 2 with that of Mukhopadhyay (1984), and Mukhopadhyay's (1984) procedure comes out as the inferior one. To facilitate this comparison, we need a specialized table (Table I) for the Studentized smaller Chi-squares, and we provide with an exact method to do just that. The Table I is also very useful for another related problem discussed in Mukhopadhyay and Hamdy (1963). In the end, we present a limited simulation study about the moderate sample-size performances of the two-stage procedures in Mukhopadhyay (1984), and Mukhopadhyay and Hamdy (1983) for k = 2. However, for a general k (≥ 3), our newly developed two-stage procedures as developed here are the only ones known to us at this time.

ISSN:0747-4946    

DOI:10.1080/07474948408836052

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor

摘要:

Techniques of Armitage (1958) for finding confidence intervals after sequential tests (SCI) are applied to curtailed binomial test boundaries. The form of the exact randomized SCI is given. We also show that the conservative confidence interval calculated as though a fixed sample size procedure had been used (FCI) remains conservative when used after stopping on a curtailed boundary. Numerical results are obtained to assess the potential gains that may be obtained by using the conservative SCI or exact SCI over the conservative FCI for these boundaries.

ISSN:0747-4946    

DOI:10.1080/07474948408836053

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor

摘要:

Suboptimal play the winner and one step ahead sequential sampling rules for the Bernoulli two armed bandit are compared with respect to several criteria.

ISSN:0747-4946    

DOI:10.1080/07474948408836054

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor

摘要:

A sequential prediction problem is considered and it is shown that the set of rules based on an adequate and transitive sequence forms an essentially complete class.

ISSN:0747-4946    

DOI:10.1080/07474948408836055

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor

ISSN:0747-4946    

DOI:10.1080/07474948408836048

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor