xqplot: Extended Quantile Plots

In gmonette/spida15: Collection of miscellaneous functions for mixed models etc. prepared for SPIDA 2009+ (development version)

Description

An easy way to see a dataset's variables at a glance. Shows uniform quantile plot for numerical varibles and barcharts for factors. Quantile plots also show a horizontal line at the position of the mean and at mean plus or minus one standard deviation.

Usage

xqplot(x, ptype = "quantile", labels = dimnames(x)[[2]], 
       ..., 
       mfrow = findmfrow(ncol(x)), 
       ask = prod(mfrow) < ncol(x) && dev.interac
       tive(), 
       mcex = 0.8, maxlab = 12, debug = F, 
       mar = F, maxvarnamelength = 20)

Arguments

x	a data frame or list of variables to plot
ptype	"quantile" (default) or "normal": kind of quantile to plot on x axis.
labels	names for each plot
...	additional arguments passed to 'plot' command
mfrow	number of rows and columns per page. If missing, an attempt is made to choose a reasonable number.
ask
mcex	character expansion factor for marginal text
maxlab	maximum number of categories to label in barcharts
debug	if TRUE, print additional information
mar	size of margins
text.cex.factor	character expansion factor for barchart labels
left.labs	determines placement of barchart labels
maxvarnamelength	maximum length of variable name without splitting on two lines.

Note

Bugs:

'mfrow' should take the total number of variables into account if they will fill more than one page so the last page is close to being full.

The current version of the function could be made much simpler and more transparent. Some code is redundant.

Author(s)

Georges Monette <[e-mail]>

Examples

require(car)
xqplot(Prestige)
xqplot(Prestige,"n") # normal quantiles

gmonette/spida15 documentation built on May 17, 2019, 7:26 a.m.