It takes a fitted
bgeva object produced by
bgeva() and plots the
component smooth functions that make it up on the scale of the linear predictor.
This function is based on
mgcv. Please see the documentation of
plot.gam() for full details.
Other graphics parameters to pass on to plotting commands, as described for
This function produces plot showing the smooth terms of a fitted semiparametric bivariate probit model. For plots
of 1-D smooths, the x axis of each plot is labelled using the name of the regressor, while the y axis is labelled as
regr is the regressor name, and
edf the estimated degrees of freedom of the smooth. As for 2-D smooths, perspective plots are produced with the x-axes labelled with the first and second variable names and the y axis
is labelled as
s(var1,var2,edf), which indicates the variables of which the term is a function and the
edf for the term.
seWithMean=TRUE, then the confidence intervals include the uncertainty about the overall mean. That is,
although each smooth is shown centred, the confidence intervals are obtained as if every other term in the model was
constrained to have average 0 (average taken over the covariate values) except for the smooth being plotted. The theoretical arguments
and simulation study of Marra and Wood (2012) suggests that
seWithMean=TRUE results in intervals with
close to nominal frequentist coverage probabilities. This option should not be used when fitting a random effect model.
The function generates plots.
The function can not deal with smooths of more than 2 variables.
Maintainer: Giampiero Marra firstname.lastname@example.org
Marra G. and Wood S.N. (2012), Coverage Properties of Confidence Intervals for Generalized Additive Model Components. Scandinavian Journal of Statistics, 39(1), 53-74.
## see examples for bgeva