The conventional entry test for a new variable in stepwise regression is not theoretically correct, but is used because the necessary exact distributions are not known. Some distributional results ...
Journal of the Royal Statistical Society. Series C (Applied Statistics), Vol. 48, No. 3 (1999), pp. 313-329 (17 pages) The number of variables in a regression model is often too large and a more ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback