Description Usage Arguments Details Value Author(s) See Also Examples
This routine performs further computations on a list of feat.rank.re
objects contained in a
mfr.obj
object (see ftrank.agg
). If only one feat.rank.re
result from feat.rank.re
is analyzed, it is easier to pass it first to ftrank.agg
(see example).Two calculations are made: 1) Computation of the pseudo mrp-value from the resampling based feature (see fs.mrpval
) and 2) adjusted p values if p values have calculated (see p.adjust
).
1 | summ.ftrank(lclas, lmod = NULL, qtl = 0.25, padjust = "fdr")
|
lclas |
mfr.obj object - See details in |
lmod |
List of models to be considered in lclas |
qtl |
Quantile - See details in |
padjust |
p value adjustement method - See details in |
The resulting list as two component: one is equal to the total number of feat.rank.re
objects (i.e one resampling experiment) and one field is a table that summarises each feat.rank.re
(as for ftrank.agg
). In the first component, each table may have a different number of columns depending if the FR method outputs p-values or not:
Original statistics calculated on the overall data.
Original feature rank calculated on the overall data.
Original feature p-value calculated on the overall data (optional).
Feature adjusted p-value by method xxx if p-value available (i.e. previous column).
Pseudo multiple resampling p-value using a given qtl
value xxx.
Average feature rank calculated from the ranks found for each training data partition.
Associated feature rank standard deviation calculated from the ranks found for each training data partition.
mfr.sum
object:
frsum |
List of tables corresponding to each |
frdef |
Summary of each |
David Enot dle@aber.ac.uk
fs.mrpval
, tidy.ftrank
, p.adjust
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | data(abr1)
y <- factor(abr1$fact$class)
x <- preproc(abr1$pos , y=y, method=c("log10","TICnorm"),add=1)[,110:500]
## Select classes 1 and 2
dat <- dat.sel1(x, y, pwise="1",mclass=NULL,pars=valipars(sampling="boot",niter=2,nreps=5))
reswelch = feat.rank.re(dat[[1]],method="fs.welch")
mfr=ftrank.agg(reswelch)
print(mfr$frdef)
frsum=summ.ftrank(mfr,lmod=1,qtl=.3)
## print the FR components
print(frsum$frdef)
## have a look at the first 5 variables
print(frsum$frsum[[1]][1:5,])
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.