Downloads & Free Reading Options - Results

Using a collection of simulated an real benchmarks, we compare Bayesian and frequentist regularization approaches under a low informative constraint when the number of variables is almost equal to the number of observations on simulated and real datasets. This comparison includes new global noninformative approaches for Bayesian variable selection built on Zellner's g-priors that are similar to Liang et al. (2008). The interest of those calibration-free proposals is discussed. The numerical experiments we present highlight the appeal of Bayesian regularization methods, when compared with non-Bayesian alternatives. They dominate frequentist methods in the sense that they provide smaller prediction errors while selecting the most relevant variables in a parsimonious way.

The Author of The Ghost Stories of an Antiquary in the preface to one of his books expressed his lively distaste for benevolent ghosts, and ghosts with nice minds. The author profoundly agrees with this sentiment of the master, and, furthermore, he abominates the 'natural' explanation, a poisonous anti-climax. So this much can be said for his tales, that those Who Return therein are animated by undiluted malevolence, and no iconoclastic materialist has been allowed to cast a doubt on their credentials as genuine apparitions. (Summary from Inside Front Dust Cover of 1928 First Edition)