SemiParBIVProbit (version 3.2-4)

plot.SemiParBIVProbit: SemiParBIVProbit plotting

Description

It takes a fitted SemiParBIVProbit object produced by SemiParBIVProbit() and plots the component smooth functions that make it up on the scale of the linear predictor.

Usage

## S3 method for class 'SemiParBIVProbit':
plot(x, eq, select, rug=TRUE, se=TRUE, se.l=1.95996, seWithMean=FALSE, 
     n=100, xlab = NULL, ylab=NULL, zlab=NULL, xlim=NULL, ylim = NULL, 
     main=NULL, trans = I, n2 = 40, theta = 30, phi = 30, too.far = 0.1, ...)

Arguments

x

A fitted SemiParBIVProbit object as produced by SemiParBIVProbit().

eq

The equation from which smooth terms should be considered for printing.

select

The smooth component to be considered for printing.

rug

If TRUE (default), then the covariate to which the plot applies is displayed as a rug plot at the foot of each plot of a 1-D smooth.

se

If TRUE (default), then Bayesian `confidence' intervals are added to 1-D plots at se.l standard errors above and below the estimate of the smooth being plotted.

se.l

Significance level.

seWithMean

If TRUE, then the component smooth is shown with confidence intervals that include the uncertainty about the overall mean. If FALSE, then the uncertainty relates purely to the centred smooth itself.

n

Number of points used for each 1-D plot.

xlab

If supplied, then this is used as the x label.

ylab

If supplied, then this is used as the y label.

zlab

If supplied, then this is used as the z label for 2-D smooths.

xlim

If supplied, then this pair of numbers are used as the x limits for 2-D smooths.

ylim

If supplied, then this pair of numbers are used as the y limits.

main

If supplied, then this is used as title.

trans

Function to apply to each smooth before plotting.

n2

Square root of number of points used to grid estimates of 2-D functions.

theta

One of the perspective plot angles.

phi

The other perspective plot angle.

too.far

When plotting 2-D smooths, if greater than 0, then it is used to determine when a location is too far from data to be plotted. The data are scaled into the unit square before deciding what to exclude, and too.far is a distance within the u

...

Other graphics parameters to pass on to plotting commands.

Value

• The function generates plots.

WARNING

The function can not deal with smooths of more than 2 variables.

Details

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 s(regr,edf) where 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. If 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 function is pretty similar to plot.gam in mgcv.

References

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 Also

AT, InfCr, SemiParBIVProbit, summary.SemiParBIVProbit, predict.SemiParBIVProbit

Examples

## see examples for SemiParBIVProbit