-
作者：Kennedy, MC; O'Hagan, A
作者单位：University of Sheffield
摘要：We consider prediction and uncertainty analysis for complex computer codes which can be run at different levels of sophistication. In particular, we wish to improve efficiency by combining expensive runs of the most complex versions of the code with relatively cheap runs from one or more simpler approximations. A Bayesian approach is described in which prior beliefs about the codes are represented in terms of Gaussian processes. An example is presented using two versions of an oil reservoir si...
-
作者：Bingham, NH
作者单位：University of London; Birkbeck University London
摘要：Both probability and statistics had developed into major fields by the end of the nineteenth century, with a sizeable body of theory and extensive application, but both disciplines achieved their existing form only this century, by making use of modern ideas. This study traces this evolution, with particular reference to the role of measure theory, from its introduction in definitive form by Lebesgue in 1902 to its successful use in the axiomatisation of probability by Kolmogorov in 1933.
-
作者：Verdinelli, I
作者单位：Carnegie Mellon University
摘要：The Bayesian theory of optimal experimental design for the normal linear model has been developed under the assumption of known variance. The insensitivity of specific design criteria to prior assumptions on the variance distribution has been demonstrated in special cases, but a general result showing the way in which Bayesian optimal designs are affected by prior information on the variance is lacking. This note proves that Bayesian designs are insensitive to information about the variance in...
-
作者：Hald, A
作者单位：University of Copenhagen
摘要：Pizzetti's work is considered as bridging the gap between Helmert and Fisher. By means of an orthonormal transformation Pizzetti decomposed the total sum of squares in the linear normal model into its constituent and independent parts, thus providing the theoretical basis for the analysis of variance in the fixed effects model. He calculated a table of the chi(2) distribution, which he used for finding probability limits for sla. He derived the estimates of the components of variance in the ra...
-
作者：Robins, JM; Wang, NS
作者单位：Harvard University; Harvard T.H. Chan School of Public Health; Texas A&M University System; Texas A&M University College Station
摘要：We derive an estimator of the asymptotic variance of both single and multiple imputation estimators. We assume a parametric imputation model but allow for non- and semiparametric analysis models. Our variance estimator, in contrast to the estimator proposed by Rubin (1987), is consistent even when the imputation and analysis models are misspecified and incompatible with one another.
-
作者：Seber, GAF; Nyangoma, SO
作者单位：University of Auckland
摘要：Ordinary residuals from linear models are useful for diagnostic purposes. Unfortunately residuals from nonlinear models do not share the same useful properties, and Cook & Tsai (1985) introduced so-called projected residuals which overcome some of the problems. After setting up some second-order asymptotics, we briefly discuss various residuals used currently for multinomial models and introduce some new diagnostics including projected residuals. Two examples, from genetics and psychology, are...
