saemix.plot.setoptions: Function setting the default options for the plots in SAEM

In saemix: Stochastic Approximation Expectation Maximization (SAEM) Algorithm

Function setting the default options for the plots in SAEM

Description

• ablinecol Color of the lines added to the plots (default: "DarkRed")

• ablinelty Type of the lines added to the plots. Defaults to 2 (dashed line)

• ablinelwd Width of the lines added to the plots (default: 2)

• ask A logical value. If TRUE, users will be prompted before each new plot. Defaults to FALSE

• cex A numerical value giving the amount by which plotting text and symbols should be magnified relative to the default. Defaults to 1 (no magnification)

• cex.axis Magnification to be used for axis annotation relative to the current setting of 'cex'. Defaults to 1 (no magnification)

• cex.main Magnification to be used for main titles relative to the current setting of 'cex'. Defaults to 1 (no magnification)

• cex.lab Magnification to be used for x and y labels relative to the current setting of 'cex'. Defaults to 1 (no magnification)

• col.fillmed For the VPC plots: color filling the prediction interval for the median. Defaults to "pink"

• col.fillpi For the VPC plots: color filling the prediction interval for the limits of the prediction interval. Defaults to "slategray1"

• col.lmed For the VPC plots: color of the line showing the median of the simulated data. Defaults to "indianred4"

• col.lobs For the VPC plots: color of the lines showing the median, 2.5 and 97.5th percentiles (for a 95% prediction interval). Defaults to "steelblue4"

• col.lpi For the VPC plots: color of the line showing the boundaries of the prediction intervals. Defaults to "slategray4"

• col.obs For the VPC plots: color used to plot the observations. Defaults to "steelblue4"

• cov.name Name of the covariate to be used in the plots. Defaults to the first covariate in the model

• cov.value Value of the covariate to be used in the plots. Defaults to NA, indicating that the median value of the covariate (for continuous covariates) or the reference category (for categorical covariates) will be used

• ilist List of indices of subjects to be included in the individual plots (defaults to all subjects)

• indiv.par a string, giving the type of the individual estimates ("map"= conditional mode, "eap"=conditional mean). Defaults to conditional mode

• lcol Main line color (default: black)

• line.smooth Type of smoothing when a smoothed line is used in the plot ("m": mean value, "l": linear regression; "s": natural splines). Several options may be combined, for instance "ls" will add both a linear regression line and a line representing the fit of a natural spline. Defaults to "s"

• lty Line type. Defaults to 1, corresponding to a straight line

• lty.lmed For the VPC plots: type of the line showing the median of the simulated data. Defaults to 2 (dashed)

• lty.obs For the VPC plots: type of the line showing the observed data. Defaults to 1

• lty.lpi For the VPC plots: type of the line showing the boundaries of the simulated data. Defaults to 2 (dashed)

• lwd Line width (default: 1)

• lwd.lmed For the VPC plots: thickness of the line showing the median of the simulated data. Defaults to 2

• lwd.obs For the VPC plots: thickness of the line showing the median and boundaries of the observed data. Defaults to 2

• lwd.lpi For the VPC plots: thickness of the line showing the boundaries of the simulated data. Defaults to 1

• par.name Name of the parameter to be used in the plots. Defaults to the first parameter in the model

• pch Symbol type. Defaults to 20, corresponding to small dots

• pcol Main symbol color (default: black)

• range Range (expressed in number of SD) over which to plot the marginal distribution. Defaults to 4, so that the random effects for the marginal distribution is taken over the range [-4 SD; 4 SD]

• res.plot Type of residual plot ("res.vs.x": scatterplot versus X, "res.vs.pred": scatterplot versus predictions, "hist": histogram, "qqplot": QQ-plot) (default: "res.vs.x")

• smooth When TRUE, smoothed lines are added in the plots of predictions versus observations (default: FALSE)

• tit Title of the graph (default: none)

• type Type of the plot (as in the R plot function. Defaults to "b", so that both lines and symbols are shown

• units Name of the predictor used in the plots (X). Defaults to the name of the first predictor in the model (saemix.data$names$predictors[1])

• vpc.bin Number of binning intervals when plotting the VPC (the (vpc.bin-1) breakpoints are taken as the empirical quantiles of the X data). Defaults to 10

• vpc.interval Size of the prediction intervals.Defaults to 0.95 for the 95\

• vpc.obs Should the observations be overlayed on the VPC plot. Defaults to TRUE

• vpc.pi Should prediction bands be computed around the median and the bounds of the prediction intervals for the VPC. Defaults to TRUE

• xlab Label for the X-axis. Defaults to the name of the X predictor followed by the unit in bracket (eg "Time (hr)")

• xlim Range for the X-axis. Defaults to NA, indicating that the range is to be set by the plot function

• xlog A logical value. If TRUE, a logarithmic scale is in use. Defaults to FALSE

• xname Name of the predictor used in the plots (X)

• ylab Label for the Y-axis. Defaults to the name of the response followed by the unit in bracket (eg "Concentration (mg/L)" (Default: none)

• ylim Range for the Y-axis. Defaults to NA, indicating that the range is to be set by the plot function

• ylog A logical value. If TRUE, a logarithmic scale is in use. Defaults to FALSE

Plotting a SaemixData object also allows the following options:

individual

if TRUE, plots separate plots for each individual, otherwise plots a spaghetti plot of all the data. Defaults to FALSE

limit

for individual plots, plots only a limited number of subjets (nmax). Defaults to TRUE

nmax

for individual plots, when limit is TRUE, the maximum number of plots to produce. Defaults to 12

sample

for individual plots, if TRUE, randomly samples nmax different subjects to plot. Defaults to FALSE (the first nmax subjects are used in the plots)

Usage

saemix.plot.setoptions(saemixObject)

Arguments

saemixObject

an object returned by the saemix function

Details

This function can be used to create a list containing the default options and arguments used by the plot functions.

A more detailed description of the options set via these lists is provided in the PDF documentation. The "replace" functions are helper functions used within the plot functions. saemix.plot.setoptions has more available options than saemix.data.setoptions since it applies to all possible plots while the latter only applies to data.

Value

A list containing the options set at their default value. This list can be stored in an object and its elements modified to provide suitable graphs.

Author(s)

Emmanuelle Comets [e-mail], Audrey Lavenu, Marc Lavielle.

References

E Comets, A Lavenu, M Lavielle M (2017). Parameter estimation in nonlinear mixed effect models using saemix, an R implementation of the SAEM algorithm. Journal of Statistical Software, 80(3):1-41.

E Kuhn, M Lavielle (2005). Maximum likelihood estimation in nonlinear mixed effects models. Computational Statistics and Data Analysis, 49(4):1020-1038.

E Comets, A Lavenu, M Lavielle (2011). SAEMIX, an R version of the SAEM algorithm. 20th meeting of the Population Approach Group in Europe, Athens, Greece, Abstr 2173.

See Also

SaemixObject,saemix, saemix.plot.data, saemix.plot.convergence, saemix.plot.llis, saemix.plot.randeff, saemix.plot.obsvspred, saemix.plot.fits, saemix.plot.parcov, saemix.plot.distpsi, saemix.plot.scatterresiduals, saemix.plot.vpc, saemix.plot.mirror

Examples

# Theophylline example, after a call to fit.saemix (see examples)
# Not run
# sopt<-saemix.plot.setoptions(saemix.fit)
# sopt$ask<-TRUE

saemix documentation built on May 29, 2024, 10:43 a.m.