-
作者：Lee, SMS; Young, GA
作者单位：University of Cambridge; University of Hong Kong
摘要：A version of the sequential probability ratio test for testing simultaneously a set of nested hypotheses is developed. This procedure is then applied to define a sequential procedure of sampling at the inner level of the two nested levels of resampling required by Monte Carlo construction of an iterated bootstrap percentile method confidence interval. The sequential resampling scheme reduces very significantly the computational demands of construction of the iterated bootstrap confidence inter...
-
作者：Tibshirani, R
摘要：We propose a new method for estimation in linear models. The 'lasso' minimizes the residual sum of squares subject to the sum of the absolute value of the coefficients being less than a constant. Because of the nature of this constraint it tends to produce some coefficients that are exactly 0 and hence gives interpretable models. Our simulation studies suggest that the lasso enjoys some of the favourable properties of both subset selection and ridge regression. It produces interpretable models...
-
作者：Atkinson, AC
摘要：Optimum experimental designs were originally developed by Kiefer, mainly for response surface models. This survey of recent developments emphasizes potential or actual usefulness. For linear models the construction of exact designs, particularly over irregular design regions, is stressed, as is the blocking of response surface designs. Other important areas include systematic designs that are robust against trend and designs for mixtures with irregular design regions: several industrial exampl...
-
作者：Hastie, T; Tibshirani, R
作者单位：University of Toronto
摘要：Fisher-Rao linear discriminant analysis (LDA) is a valuable tool for multigroup classification. LDA is equivalent to maximum likelihood classification assuming Gaussian distributions for each class. In this paper, we fit Gaussian mixtures to each class to facilitate effective classification in non-normal settings, especially when the classes are clustered. Low dimensional views are an important by-product of LDA - our new techniques inherit this feature. We can control the within-class spread ...
-
作者：Jones, B; Sebastiani, P; Lewis, SM; Giovagnoli, A; Torsney, B; Gilmour, SG; Abraham, I; Firth, D; Atkinson, AC; Chan, PLY; Muller, WG; Bogacka, B; Box, G; Donev, AN; Fang, KT; Hickernell, FJ; Govaerts, B; Haines, LM; Schwabe, R; Settimi, R; Titterington, DM; Atkinson, AC; Bates, RA; Buck, RJ; Riccomagno, E; Wynn, HP
作者单位：City St Georges, University of London; University of Southampton; University of Udine; University of Pavia; University of Glasgow; University of Reading; University of Oxford; University of London; London School Economics & Political Science; University of Hong Kong; Poznan University of Life Sciences; University of Wisconsin System; University of Wisconsin Madison; Hong Kong Baptist University; Solvay SA; Universite Catholique Louvain; University of Kwazulu Natal; Free University of Berlin; University of Perugia
-
作者：Festing, MFW; Lovell, DP
摘要：More than 50 million animals are used in biomedical research in the world each year. It is highly desirable that this number is reduced both for ethical and for economic reasons. Better experimental design could lead to the use of fewer animals and improve the repeatability of animal experiments so that alternative methods would be easier to validate. Screening experiments aimed at identifying rodent carcinogens would be more powerful if more than one strain of mice and/or rats were used. Atte...
