grasp.model {grasper} | R Documentation |
Creates a model from the given variables
grasp.model(gr.Yi, sX, family, trace, df, calcdf
gr.Yi |
Selected responses to be used in the form of a vector with column numbers |
sX |
Selected predictors to be used in the form of a vector with column numbers |
family |
Family to be used for the model, see details |
trace |
Verbose |
df |
Degrees of freedom to be used for each variable |
calcdf |
If TRUE, the df are calculated by the gam function |
This function creates the model from the given variables. The family is important. You can choose between gaussian
, binomial
, quasibinomial
and quasipoisson
. See family
for details. To create the model, the function uses the gam
function provided by the mgcv package. The grasp.GUI
function provides method for selecting the appropriate family based on which type of variable you have.Grasp.model has to be used before the stepwise selection procedure (see grasp.step.gam
.
Fabien Fivaz fabien.fivaz@bluewin.ch. Ported to R from GRASP http://www.cscf.ch/grasp/ for S-Plus written by A. Lehmann, J.R. Leathwich and J. McC Overton.
grasp.step.gam
, gam
and grasp.pred