mfp.object: Multiple Fractional Polynomial Model Object
Description
Objects returned by fitting fractional polynomial model objects.
These are objects representing fitted mfp models.
Class mfp inherits from either glm or coxph depending on the type of model fitted.
Arguments
Value
In addition to the standard glm/coxph components the following components are
included in a mfp object.
xthe final FP transformations that are contained in the design matrix x.
The predictor "z" with 4 df would have corresponding columns "z.1" and
"z.2" in x.
powersa matrix containing the best FP powers for each predictor. If a predictor
has less than two powers a NA will fill the appropriate cell of the
matrix.
pvaluesa matrix containing the P-values from the closed tests. Briefly p.null is
the P-value for the test of inclusion (see mfp), p.lin corresponds to
the test of nonlinearity and p.FP the test of simplification. The best m=1
power (power2) and best m=2 powers (power4.1 and power4.2) are also given.
scaleall predictors are shifted and rescaled before being power transformed if
nonpositive values are encountered or the range of the predictor is
reasonably large. If x' would be used instead of x where
x' = (x+a)/b
the parameters a (shift) and b (scale) are contained in the matrix scale.
df.initiala vector containing the degrees of freedom allocated to each predictor.
df.finala vector containing the degrees of freedom of each predictor at
convergence of the backfitting algorithm.
devthe deviance of the final model.
dev.linthe deviance of the model that has every predictor included with 1 df
(i.e. linear).
dev.nullthe deviance of the null model.
fptablethe table of the final fp transformations.
formulathe proposed formula for a call of glm/coxph.
fitthe fitted glm/coxph model using the proposed formula. This component can be used for prediction, etc.