-
作者：BARNDORFFNIELSEN, OE; JAMES, IR; LEIGH, GM
作者单位：University of Western Australia; Commonwealth Scientific & Industrial Research Organisation (CSIRO)
摘要：A semiparametric estimator of natural mortality, introduced by Hearn, Sandland and Hampton (1987), was studied by Leigh (1988), who noted that its asymptotic variance may sometimes be infinite. We discuss the nonnormal asymptotic behaviour that can occur in this case, and remark on difficulties that it causes in practice.
-
作者：MUKERJEE, R; SENGUPTA, S
作者单位：University of Calcutta
摘要：Restricting attention to fixed size sampling designs and linear unbiased estimators of a finite population total, we give methods for findings estimators with minimum model expected variance and the optimal strategy under a general correlated superpopulation model. Some techniques popular in the theory of optimal experiments help in the derivation. Several earlier optimality results are deduced as special cases.
-
作者：VARDI, Y
摘要：This paper derives the nonparametric maximum likelihood extimate for a lifetime distribution G, under the following ''multiplicative-censorship'' model: X1,...,Xm are complete uncensored observations from G, and Y1,..., Yn are incomplete observations from G. The incompleteness of the Yi''s is assumed to come from the following censoring mechanism. For each Yi there exists and unobserved parent-observation'' Zi, distributed according to G, and Yi is the product of Zi and an independent uniform ...
-
作者：WEI, LJ; SMYTHE, RT; MEHTA, CR
作者单位：George Washington University; Harvard University; Harvard University Medical Affiliates; Dana-Farber Cancer Institute; Harvard Medical School
-
作者：GUPTA, S
摘要：The work of Pearce is studied in relation to the recent research on designs for control versus test comparisons. A condition for a nonorthogonal design to be most efficient is given, and a lower bound to the average variance of supplemented contrasts is obtained.
-
作者：TAYLOR, CC
-
作者：WHITTEMORE, AS; HALPERN, J
摘要：We present a test for the hypothesis that odds-ratios are constant across two or more diseases, using matched case-control data. The test requires less computing than the usual conditional likelihood score test (Prentice, 1979) because it excludes the control covariates. We give a practical algorithm to compute the test statistic and discuss its power.
-
作者：AGRAWAL, MC; JAIN, N
作者单位：University of Delhi
摘要：A product estimator fo the population mean which is predictive in character is proposed. Although optimal in the sense of unbiasedness and minimum mean squared error under a certain superpopulation model, the estimator is frequently less biased than its competitors under a sampling theoretic approach.
-
作者：STRUTHERS, CA; FAREWELL, VT
摘要：The use of a mixture model to estimate the probability of developing AIDS after HIV infection and the distribution of incubation time is investigated. The possibility of left truncation and an uncertain time of infection is incorporated into the analysis. The efficiency losses associated with uncertain infection times are investigated. An example is given.
-
作者：BURNETT, R; KREWSKI, D; BLEUER, S
摘要：The standard statistical test for assessing adverse effects in rodent tumourigenicity experiments is based on a time-adjusted analysis of lethal and incidental tumours. This test is shown to arise as the sum of two score statistics for fatal and incidental lesions respectively, the former optimal against extreme value hazard alternatives and the latter optimal against logistic prevalence alternatives in terms of asymptotic relative efficiency in the Pitman sense. An efficiency robust test is d...
