摘要:

We consider the problem of identifying a population with one of the two given populations. A sample is given from the population to be identified and from the other two populations we can sample either sequentially or non-sequentially. The populations are considered to be normal, differing only in their unknown means, and the common variance may be known or unknown. In the non-sequential case the HPKE (Holley, Preston, Kemperman, and Edwards) inequality yields partial monotonicity for the error probabilities of the best fixed sample invariant test. In the sequential case a truncated invariant SPRT is proposed as a solution, since the untruncated SPRT does not terminate with probability one Further a simple technique yields good bounds on the error probabi!ities Numerical comparisons with the fixed sample invariant test show that even though we are not in the i.i.d. situation, the truncated SPRT achieves subsantial savings. Some other numerical studies relating to the performance of this SPRT are also made here.

ISSN:0747-4946    

DOI:10.1080/07474948408836057

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor

摘要:

Sequential procedures are developed for detecting deviations in the parameters of a time-dependent autoregressive model from specified targets. This model can be formulated so that its parameters represent typically relevant aspects, e.g. the mean or variance, of a time series, and hence is useful for monitoring these aspects of autoregressive series. The procedures are based on scores-type statistics aimed at detecting arbitrary shifts in the parameters, and are capable of addressing any subset of the model parameters, so that any combination of these aspects may be monitored. The procedures' false signal rates are controlled, and their reaction quickness under local shift alternatives studied, via invariance principles obtained by exploiting the martingale structure of the scores vector. Non-local shifts in the mean and variance are studied via simulation

ISSN:0747-4946    

DOI:10.1080/07474948408836058

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor

ISSN:0747-4946    

DOI:10.1080/07474948408836059

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor

摘要:

The effect of the contamination of the form Fθ,ε=(1-ε)Fθ+ε δO,0≤ε≤1,on the Chow-Robbins sequential confidence interval of the length ≤2d is studied. If ε is negligible with respect to dl then the contamination has no effect on the asymptotic coverage probability and on the behavior of the stopping rule, The contaminationproportional to d changes the coverage probability and causes a bias in the asymptotic distribution of the stopping rule, The results are illustrated by a Monte-Carlo study

ISSN:0747-4946    

DOI:10.1080/07474948408836060

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor

ISSN:0747-4946    

DOI:10.1080/07474948408836056

出版商:Marcel Dekker, Inc.    

年代:1984

数据来源: Taylor