摘要:

In this paper a muIt,ivariat.e EWMA chart for time series is introduced. In principle, it is a generalization of the control scheme of Lowry et al. (I992) for multivarite indendent observations.

ISSN:0747-4946    

DOI:10.1080/07474949708836378

出版商:Marcel Dekker, Inc.    

年代:1997

数据来源: Taylor

摘要:

Since exact formulae for Bayes stopping tirncs in sequential analysis are often difficult to derive.Bickle and Yahav considered an attractive large sample approximation known as " Asymptotically pointwise optimal" (A.P.O.) rulp. The (A.P.O.) rule. for the problem of cst the mean of a multivariate normal distribution has been treated by some authors but only for the case where the covariance matrix is scalar times a known matrix.In this paper, A.P.O. rules are given for a general exponeritial fitmily fror which the results for milltic,aria.tr norrnal mean vector with c-ompli.tely unknow covariance matrix are obtained. We also consider a multinomial distribution case.

ISSN:0747-4946    

DOI:10.1080/07474949708836379

出版商:Marcel Dekker, Inc.    

年代:1997

数据来源: Taylor

摘要:

In normal populations,Healy (1956) and Takada(1988) developed two-stage procedures for inference on the mean vector,with a confidence region which has a specified maximum diameter or with a point estimation which has a specified risk.Those procedures are proved to remain valid even under a compounded normal distribution,whose conditional distribution given the tail sigma-algebra is normal with unknown mean μ and covariance matrix ∑/rV. where r is a constant and V is a random variable.It is shown that Takada's procedure is asymptotically efficienct in terms of expected sample size for the compounded normal distribution. Super efficiency is observed in Healy s procedure when V is distributed aswhere νis known.In other words.Healy's procedure with estimated ∑ performs better than the fixed-sample procedure when ∑is known.A discussion on super efficiency is given.

ISSN:0747-4946    

DOI:10.1080/07474949708836380

出版商:Marcel Dekker, Inc.    

年代:1997

数据来源: Taylor

摘要:

In Lemma 4.1 of Kiefer and Weiss (1957) the duality between Bayesian tests and optimal sequential tests with prescribed upper bounds for the risks of first and second kind is shown. We provide a generalization of this lemma for more than two hypotheses and a more general class of decision problems. The result could be used for instance, if the optimal sequential tests are to be constructed by means of iteratively applied backward induction.

ISSN:0747-4946    

DOI:10.1080/07474949708836381

出版商:Marcel Dekker, Inc.    

年代:1997

数据来源: Taylor

摘要:

Let X1,X2,... be a sequential experimentWe give a condition for the existence of a corresponding experimentwhere Ziare independent and the experiments {Xi} and {Zi} are sequentially equivalent. An application for testing problems is given.

ISSN:0747-4946    

DOI:10.1080/07474949708836382

出版商:Marcel Dekker, Inc.    

年代:1997

数据来源: Taylor

ISSN:0747-4946    

DOI:10.1080/07474949708836377

出版商:Marcel Dekker, Inc.    

年代:1997

数据来源: Taylor