BoostMLR (version 1.0.3)

plotVIMP: Variable Importance (VIMP) plot

Description

Barplot displaying variable importance for the main effect.

Usage

plotVIMP(vimp_Object,
         xvar.names = NULL,
         cex.xlab = NULL,
         ymaxlim = 0,
         yminlim = 0,
         main = "Variable Importance (%)",
         col = grey(0.8),
         cex.lab = 1.5,
         ylbl = NULL,
         legend_placement = NULL,
         plot.it = TRUE,
         path_saveplot = NULL,
         Verbose = TRUE)

Arguments

vimp_Object

List with number of elements equal to the number of response variables.

xvar.names

Names of the covariates. If NULL, names will be pulled from vimp_Object.

cex.xlab

Magnification of the names of the covariates for the barplot.

ymaxlim

By default, we use the range of the vimp values for the barplot limit on the y-axis. If one wants to extend the limit, add the amount with which the limit will extend above the x-axis.

yminlim

Similar to ymaxlim, this will add the amount with which the limit will extend below the x-axis.

main

Main title for the plot.

col

Color of the plot.

cex.lab

Magnification of the x and y lables.

ylbl

Label for the y-axis.

legend_placement

Do you want name of the covariates on top of the each barplot? If so, use default setting; else set value on the negative direction of y-axis which arrange covariate name beneath the barplot.

plot.it

Should the VIMP plot be displayed?

path_saveplot

Provide the location where plot should be saved. By default the plot will be saved at temporary folder.

Verbose

Display the path where the plot is saved?

Details

Barplot displaying VIMP for each response. Barplot will be save as pdf file in the working directory.

Examples

# NOT RUN {
##-----------------------------------------------------------------
## VIMP plot for multivariate longitudinal response
##-----------------------------------------------------------------

# Simulate data involves 3 response and 4 covariates

dta <- simLong(n = 100, ntest = 100 ,N = 5, rho =.80, model = 1, q_x = 0, 
                                  q_y = 0,type = "corCompSym")
dtaL <- dta$dtaL
trn <- dta$trn
# Boosting call: Raw values of covariates, B-spline for time, 
# no shrinkage, no estimate of rho and phi

boost.grow <- BoostMLR(x = dtaL$features[trn,], tm = dtaL$time[trn], 
                      id = dtaL$id[trn], y = dtaL$y[trn,], M = 100, VarFlag = FALSE)

boost.pred <- predictBoostMLR(Object = boost.grow, x = dtaL$features[-trn,], 
                               tm = dtaL$time[-trn], id = dtaL$id[-trn], 
                               y = dtaL$y[-trn,], importance = TRUE)

# Plot VIMP
plotVIMP(vimp_Object = boost.pred$vimp,ymaxlim = 20,plot.it = FALSE)
# }