data
argument is a bigdata bdFrame (see the DETAILS).
qq(formula, f.value = ppoints, ...)The following arguments have special meaning within this function. The common meanings for these and all other arguments are listed separately under
trellis.args
.
y ~ x | g1 * g2 * ...
however the given variables
g1
,
g2
,
...
may be omitted.
x
is the numeric response,
and
y
is a numeric, shingle
or factor grouping vector.
y
must have exactly two levels
which divide the
x
values into two groups.
Quantiles from the first of these groups are plotted on the horizontal axis;
corresponding quantiles from the second group are plotted on the vertical axis.
(
y
will be coerced into a factor or shingle,
if it is not already such).
function(n) c(.01, .05, seq(.1,.9,.1), .95, .99)
trellis
,
which is automatically plotted by
print.trellis
.
If the
data
argument is a bdFrame then approximate quantiles
(based on binning the data into 1000 equi-spaced bins)
are computed and plotted.
# compare bass heights to tenor heights bass2 <- singer$height[singer$voice.part == "Bass 2"] tenor1 <- singer$height[singer$voice.part == "Tenor 1"] qq(which ~ data, data = make.groups(bass2,tenor1), aspect = 1, ylab = "Tenor 1 Height (inches)", xlab = "Bass 2 Height (inches)") # alternative to the above qq(voice.part ~ height, aspect = 1, data = singer, subset = voice.part == "Bass 2" | voice.part == "Tenor 1", ylab = "Tenor 1 Height (inches)", xlab = "Bass 2 Height (inches)") # compare payoffs in 2 periods of lottery drawings qq(which ~ data, data = make.groups(lottery.payoff, lottery2.payoff))