Optimal Properties of One-Step Variable Selection in Regression
Butler, Ronald W.
Ann. Statist., Tome 11 (1983) no. 1, p. 219-224 / Harvested from Project Euclid
Text on Project Euclid
PDF
Alt PDF
We consider the selection of the best subset of independent variables of a fixed size for possible inclusion in a regression model. The classical procedures (largest $\mathbb{R}^2$ to enter) are shown to be uniformly invariant Bayes in the sense of Paulson (1952) and Kudo (1956).
Publié le : 1983-03-14
Classification:  Variable selection,  Bayesian decision rule,  invariance,  regression analysis,  62C25,  62C10,  62J05,  62H30 
@article{1176346072,
     author = {Butler, Ronald W.},
     title = {Optimal Properties of One-Step Variable Selection in Regression},
     journal = {Ann. Statist.},
     volume = {11},
     number = {1},
     year = {1983},
     pages = { 219-224},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176346072}
}

Butler, Ronald W. Optimal Properties of One-Step Variable Selection in Regression. Ann. Statist., Tome 11 (1983) no. 1, pp.  219-224. http://gdmltest.u-ga.fr/item/1176346072/