plot.DALSM {DALSM}

Plot visual information on a DALSM.object

Description

Visualize the estimated additive terms and error density corresponding to a Double Additive Location-Scale Model (DALSM) object.

Usage

## S3 method for class 'DALSM'
plot(x,
       mfrow.loc=NULL, mfrow.disp=NULL,
       nx=101, equal.ylims=TRUE, true.loc=NULL,true.disp=NULL, ci.level=NULL,
       error.lim = NULL, add.residuals=FALSE, true.derr=NULL, new.dev=TRUE, ...)

Arguments

x

a DALSM.object.

mfrow.loc

(optional) window layout to plot the additive terms for location.

mfrow.disp

(optional) window layout to plot the additive terms for dispersion.

nx

(optional) number of points to make the plots for the fitted additive terms (default: 101).

equal.ylims

logical indicating if the same y-limits must be used when plotting the fitted additive terms (default: TRUE).

true.loc

(optional) list of functions to be superposed to the corresponding estimated additive terms in the location submodel (default: NULL).

true.disp

(optional) list of functions to be superposed to the corresponding estimated additive terms in the dispersion submodel (default: NULL).

ci.level

(optional) nominal level for the plotted pointwise credible intervals (default: x$ci.level).

error.lim

(optional) plotting interval for the estimated standardized error density in the DALSM model (default: support of the fitted standardized error density).

add.residuals

logical requesting to add the (possibly censored) standardized residuals to the plot of the fitted standardized error density (default: FALSE).

true.derr

(optional) density function to superpose to the estimated standardized error density when plotting (default: NULL).

new.dev

(optional) logical indicating whether a new plotting device must be opened for each graph (default: TRUE).

...

additional generic plotting arguments.

Details

Plot the fitted additive terms and the estimated standardized error density contained in the DALSM.object x.

Value

In addition to the plots, an invisible list containing the following is returned:

• J1 : ⁠ ⁠number of additive terms in the location sub-model.

• x.loc : ⁠ ⁠a nx by J1 matrix containing a regular grid of nx covariate values where the corresponding additive term in location is evaluated.

• f.loc : ⁠ ⁠a nx by J1 matrix containing the J1 fitted location additive terms evaluated at x.loc.

• se.loc : ⁠ ⁠a nx by J1 matrix containing the the pointwise standard errors of the fitted location additive terms evaluated at x.loc.

• J2 : ⁠ ⁠number of additive terms in the dispersion sub-model.

• x.disp : ⁠ ⁠a nx by J2 matrix containing a regular grid of nx covariate values where the corresponding additive term in dispersion is evaluated.

• f.disp : ⁠ ⁠a nx by J2 matrix containing the J2 fitted dispersion additive terms evaluated at x.disp.

• se.disp : ⁠ ⁠a nx by J2 matrix containing the pointwise standard errors of the fitted dispersion additive terms evaluated at x.disp.

Author(s)

Philippe Lambert [e-mail]

References

Lambert, P. (2021). Fast Bayesian inference using Laplace approximations in nonparametric double additive location-scale models with right- and interval-censored data. Computational Statistics and Data Analysis, 161: 107250. <doi:10.1016/j.csda.2021.107250>

See Also

DALSM, DALSM.object, print.DALSM

Examples

require(DALSM)
data(DALSM_IncomeData)
resp = DALSM_IncomeData[,1:2]
fit = DALSM(y=resp,
            formula1 = ~twoincomes+s(age)+s(eduyrs),
            formula2 = ~twoincomes+s(age)+s(eduyrs),
            data = DALSM_IncomeData)
plot(fit)