作者：POLI, I; JONES, RD
作者单位：United States Department of Energy (DOE); Los Alamos National Laboratory
摘要：In this article we introduce a neural net designed for nonlinear statistical prediction. The net is based on a stochastic model featuring a multilayer feedforward architecture with random connections between units and noisy response functions. A Bayesian inferential procedure for this model, based on the Kalman filter, is derived. The resulting learning algorithm generalizes the so-called one-dimensional Newton method, an updating algorithm currently popular in the neural net literature. A num...
作者：CROUX, C; ROUSSEEUW, PJ; HOSSJER, O
作者单位：University of Antwerp; Lund University
摘要：In this article we introduce a new type of positive-breakdown regression method, called a generalized S-estimator (or GS-estimator), based on the minimization of a generalized M-estimator of residual scale. We compare the class of GS-estimators with the usual S-estimators, including least median of squares. It turns out that GS-estimators attain a much higher efficiency than S-estimators, at the cost of a slightly increased worst-case bias. We investigate the breakdown point, the maxbias curve...
作者：ONEILL, TJ
作者单位：Australian National University
摘要：The logistic regression classification method uses parameter estimates that are the solution of an estimating equation. This article derives a convenient expression for the bias of a vector estimator defined by estimating equations. The expression and the results of O'Neill are used to derive the bias and the error or misclassification rate of logistic regression classification in two examples where the assumed model for logistic regression does not hold. Logistic regression classification is ...
