摘要:

We derive two-sided group sequential tests for normal responses with known variance which minimise expected sample size; minimisation is at a single value of the normal mean or integrated with respect to a normal density. In deriving these tests we use a method analogous to that used by Eales and Jennison (1992) to compute optimal one-sided tests. We present a comparison of our optimal tests with existing tests.

ISSN:0747-4946    

DOI:10.1080/07474949508836337

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

We reconsider the problem of constructing interval estimate of fixed proportional accuracy as introduced in Bose and Boukai (1993b). We discuss two approaches to make their fully sequential procedure operationally more attractive. First, we propose to implement their sequential procedure in smaller pieces along the lines of Mukhopadhyay and Sen (1993). Second, we perform purely sequential sampling part of the way, followed by appropriate batch sampling. In either case we provide the second-order expansion of the coverage probability and these compare well with that of the full sequential procedure. Along the way we also relax some of the technical restrictions of Bose and Boukai (1993b).

ISSN:0747-4946    

DOI:10.1080/07474949508836338

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

Letdenote the class of natural exponential family of distributions having power variance function. An appropriate sequential procedure for estimating the mean μ ofunder a combined loss ofweightedestimation error and sampling cost is proposed. The usual estimator, namely, where t is an appropriate stopping time, is shown to be biased. To remedy this deficiency, we propose a class of estimators to correct for the bias. The asymptotic properties of the suggested estimators are provided. In particular, one realizes the tradeoff between the bias reduction and the reduced regret compared to the ususal estimator.

ISSN:0747-4946    

DOI:10.1080/07474949508836339

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

We go back to the well known Stein two stage confidence interval and find global upper bounds on the ratio of the expected sample size for appropriate nonnormal data and normal data. The first stage sample size is taken to be 2 and the nonnormal distributions considered are the bounded uniform, t with 3 and 5 degree of freedom, the Double Exponential, the Logistic, and the Von Mises extreme value density. Notice the last distribution is asymmetric.

ISSN:0747-4946    

DOI:10.1080/07474949508836340

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

For the sequential multinomial sampling, Koike gives a sufficient condition for a stopping rule to be closed. We show that his result cannot be improved in a sense, but the condition is not necessary.

ISSN:0747-4946    

DOI:10.1080/07474949508836341

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

We give a generalization to the case ofmhypotheses of a theorem of Lai and derive an asymptotic optimality property of the Wald sequential test for general, possibly dependent, random variables with respect to therthmoment of observation time, under somer-quick convergence condition. We also extend the definition of this convergence and give an application to sequential analysis.

ISSN:0747-4946    

DOI:10.1080/07474949508836342

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

In estimating the variance of a normal distribution when the loss function is essentially squared error, a sequential version of Stein's estimator is used to show the existence of sequential estimators which are better both in risk (expected loss) and sample size than the usual estimator of a given fixed sample size.

ISSN:0747-4946    

DOI:10.1080/07474949508836343

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

Letξ1,ξ2,…be iid random variables such thatP(ξ1= 1) = p, [ILM001] and setS0= 0, Sn, = ξ1+…+ξn, n ≥ 1. Define the cusum (cumulative sum) stopping time, wherehis a positive integer. We find asymptotic (ash → ∞) distributions of suitably normalized functions ofN1(h)when(a)q > p(exponential) and(b) p = q. As an application we define a cusum procedure for detecting changes in probabilities of a multi—component process, and using (a) we obtain an asymptotic approximation for the average run length (ARL) of the given procedure. Moreover, under some suitable conditions we generalize the preceding asymptotic results for any sequence of iid random variablesX1, X2,…with meanμ≤0.

ISSN:0747-4946    

DOI:10.1080/07474949508836344

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

ISSN:0747-4946    

DOI:10.1080/07474949508836345

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

ISSN:0747-4946    

DOI:10.1080/07474949508836336

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor