bootMtrPairs and associated methods.The object class returned by boot.mtr.pairwise and associated
print, summary, and plotting classes.
# S3 method for bootMtrPairs
print(x, ...) # S3 method for bootMtrPairs
plot(x, ...)
# S3 method for bootMtrPairs
barplot(height, ylab=NULL, names.arg=height$labels,
crit.vals=TRUE, lty.crit=c(2,1), col.crit=2, lwd.crit=c(1,2), ylim=NULL, ... )
# S3 method for bootMtrPairs
summary(object, p.adjust="none", ...)
# S3 method for summary.bootMtrPairs
print(x, digits=3, ...,
signif.stars = getOption("show.signif.stars"),
signif.legend=signif.stars)
The print method returns the object, invisibly.
The plot and barplot methods return the values at the midpoint of each bar.
The summary method returns an object of class "summary.bootMtrPairs" which
is a list with members:
vector of values calculated by msd or pdchisq
character vector of labels, by default taken from x
vector of probabilities supplied and used for quantiles
matrix of quantiles. Each row corresponds to a probability
in probs and each column to an individual data point.
p-values estimated as the observed proportion of
simulated values exceeding the calculated values t0.
Character value containing the name of the p-value adjustment method used.
p-values adjusted using the given p-value adjustment method
specified by p.adjust.
Number of bootstrap replicates used.
The sampling method used by the parametric bootstrap.
The statistic subjected to bootstrapping. Either "MSD" or "PWchi2"
for msd or pdchisq, respectively.
If keep == TRUE, the individual bootstrap replicates
generated by msd or pdchisq.
t is set to NA if keep == FALSE
.
An R object. For print.bootMtrPairs and plot.bootMtrPairs, an object
of class "bootMtrPairs". For print.summary.bootMtrPairs, an object
of class "summary.bootMtrPairs".
An object of class "bootMtrPairs".
An object of class "bootMtrPairs".
Multiple correction method for calculated p-values, passed to
p.adjust.
Label for vertical axis, passed to barplot
Labels for individual bars in bar plot, passed to barplot. If names(height)
is NULL, bars are numbered.
If TRUE, individual critical values based on observation-specific
bootstrap quantiles are added to the plot. These are taken from critical.values
in the supplied bootMtrPairs object.
Vectors of line style parameters for plotted critical values, passed to
segments. Recycled to the length of critical.values
in the supplied bootMtrPairs object.
Limits for plot y range, passed to barplot. The default
ensures that the plotted bars and (if crit.vals=TRUE) the critical
values are included in the figure region.
integer; passed to print. The minimum number of
significant digits to be printed in values. Change to NULL for default.
logical; if TRUE, P-values are additionally encoded
visually as ‘significance stars’ in order to help scanning of
long coefficient tables. Defaults to the show.signif.stars
slot of options.
logical; if TRUE, a legend for the ‘significance
stars’ is printed provided signif.stars == TRUE.
Parameters passed to other methods.
S. L. R. Ellison s.ellison@lgcgroup.com
The default plot method is an alias for the barplot method.
For the plot methods, quantiles for each point are taken directly from the quantiles
calulated by boot.mtr.pairwise and retained in the returned object.
For the summary method, p-values are initially calculated as the observed
proportion of simulated values exceeding the MSD value calculated by msd. The
summary method additionally returns p-values after adjustment
for multiple comparisons using the adjustment method specified.
The print method for the summary.bootMtrPairs object prints the summary as a data
frame with columns for the calculated values (re-titled for the particular statstic),
data-specific upper quantiles (one column for each probability supplied to bootMtrPairs
and the p-values after adjustment for multiple comparisons based on the
proportion of simulated values exceeding the observed MSD.
Where that proportion is zero, the summary replaces the
raw zero proportion with 1/B, corrects that proportion using the requested
adjustment method, and reports the p-value as less than ("<") the resulting
adjusted value.
msd, qmsd, pdchisq
if (FALSE) {
data(Pb)
pwch.Pb<-pdchisq(Pb$value, Pb$u) # Uses individual standard uncertainties
set.seed(1023)
boot.Pb <- boot.mtr.pairwise(pwch.Pb)
summary(boot.Pb)
# The default summary gives individual observation p-values. To
# avoid over-interpretation for the study as a whole,
# apply a sensible p-value adjustment:
summary(boot.Pb, p.adjust="holm")
plot(boot.Pb, crit=TRUE)
}
Run the code above in your browser using DataLab