This function implements the fence procedure to find the best generalised linear model.
glmfence(mf, cstar, nvmax, adaptive = TRUE, trace = TRUE, ...)
an object of class glm
specifying the full model.
the boundary of the fence, typically found through bootstrapping.
the maximum number of variables that will be be considered in the model.
logical. If TRUE
the boundary of the fence is
given by cstar. Otherwise, it the original (non-adaptive) fence
is performed where the boundary is cstar*hat(sigma)_M,tildeM.
logical. If TRUE
the function prints out its
progress as it iterates up through the dimensions.
further arguments (currently unused)
Jiming Jiang, Thuan Nguyen, J. Sunil Rao, A simplified adaptive fence procedure, Statistics & Probability Letters, Volume 79, Issue 5, 1 March 2009, Pages 625-629, http://dx.doi.org/10.1016/j.spl.2008.10.014.