-
作者：MAZUMDAR, M; JEFFERSON, TR
摘要：The duality theory of geometric programming and the special methods available for solution can be used to obtain maximum likelihood estimates for multinomial probabilities. Three examples are given [for gene frequency, compounded probabilities and multinomial distribution estimation].
-
作者：LAUDER, IJ
摘要：Kernel models are developed and applied to the problem of [human] differential diagnosis via the diagnostic paradigm. They are shown to be a useful method both in their own right and when used in conjunction with logistic analysis.
-
作者：WORSLEY, KJ
摘要：A simple iterative procedure is found for the exact null and alternative distributions of likelihood ratio, cumulative sum and related statistics for testing for a change in probability of a sequence of independent binomial random variables. The likelihood ratio test is slightly less powerful than the cumulative sum test near the middle of the sequence, but that the likelihood ratio test is much more powerful near the ends. An example from epidemiology is used. Under certain conditions the cum...
-
作者：BECKER, N; HOPPER, JL
作者单位：University of Melbourne; Melbourne Health; Royal Melbourne Hospital
摘要：Procedures for making inference about both the mean potential that an infective has for infecting a given susceptible from the same household, and the mean potential he has for infecting a given susceptible from another household are proposed. These estimation procedures provide a check for the assumption that the disease spreads homogeneously through the community, that is the assumption of homogeneous mixing. They also provide a means of checking whether or not the spread of infection within...
-
作者：CHAN, NN; MAK, TK
作者单位：University of Hong Kong
-
作者：PETTITT, AN
摘要：An approximation to marginal rank likelihoods is given and used to make inferences for the linear regression model with censored data. The approximations involve linear rank statistics, adapted for censored data, and estimates of their variance. Various explicit scores are given for right censored data and logistic scores for doubly censored data. Inferences using the approximate rank analysis are compared with inferences using fully and partially parametric techniques on 2 data sets. [An exam...
-
作者：SAW, JG
摘要：A unit vector with m elements may be thought of as representing a direction in a space of m dimensions. Pairs of dependent unit vectors having prescribed marginal distributions are presented. Given 2 sets of unit vectors, an analysis for correlation between the 2 sets is proposed. The analysis presented is distribution free under a broad class of null distributions and thereby improves on an analysis recently suggested by Stephens (1979).
-
作者：HSIEH, FY; CROWLEY, J; TORMEY, DC
作者单位：Fred Hutchinson Cancer Center; University of Wisconsin System; University of Wisconsin Madison
-
作者：JONES, DA
-
作者：FISHER, NI; LEE, AJ
作者单位：University of Auckland
摘要：A coefficient to measure association between angular variables is discussed, its asymptotic distribution found, and its properties developed. Comparisons with other statistics in current use are made, and some examples given [including one on blood pressure].
