Linear Statistical Inference and Its Applications.
Cites background or methods from "Linear Statistical Inference and It..."
...7) is usually obtained by differentiating the log likelihood function, see Section 5a of Rao (1973), although in the context of this paper we might prefer to use the parametric bootstrap estimate of a, e....
...By now it should be clear that we can use any random variable R(y, F) to measure accuracy, not just (4.1) or (4.6), and then estimate EFIR(y, F)} by its bootstrap value EpIR(y*, F1) }- b=1 R(y*(b), F)/B. Similarly we can estimate EFR(y, F)2 by EpR(y*, F)2, etc. Efron (1983) considers the prediction problem, in which a training set of data is used to construct a prediction rule....
...R. Tibshirani is a Postdoctoral Fellow in the Department of Preventive Medicine and Biostatistics, Faculty of Medicine, University of Toronto, McMurrick Building, Toronto, Ontario, M5S 1A8, Canada. particularly Efron (1982a). Some of the discussion here is abridged from Efron and Gong (1983) and also from Efron (1984)....