R/summary_functions.R
summary.sperroresterror.Rd
summary.sperroresterror
calculates mean, standard deviation,
median etc. of the calculated error measures at the specified level
(overall, repetition, or fold). summary.sperrorestreperror
does the same
with the pooled error, at the overall or repetition level.
# S3 method for sperroresterror summary(object, level = 0, pooled = TRUE, na.rm = TRUE, ...)
object 


level  Level at which errors are summarized: 0: overall (i.e. across all repetitions); 1: repetition; 2: fold 
pooled  If 
na.rm  Remove 
...  additional arguments (currently ignored) 
Depending on the level of aggregation, a list
or data.frame
with
mean, and at level 0 also standard deviation, median and IQR of the error
measures.
Let's use an example to explain the error_rep
argument. E.g.,
assume we are using 100repeated 10fold crossvalidation. If error_rep = TRUE
(default), the mean and standard deviation calculated when
summarizing at level = 0
are calculated across the error estimates
obtained for each of the 100*10 = 1000
folds. If error_rep = FALSE
,
mean and standard deviation are calculated across the 100
repetitions,
using the weighted average of the foldlevel errors to calculate an error
value for the entire sample. This will essentially not affect the mean
value but of course the standard deviation of the error.
error_rep = FALSE
is not recommended, it is mainly for testing purposes;
when the test sets are small (as in leaveoneout crossvalidation, in the
extreme case), consider running sperrorest with error_rep = TRUE
and
examine only the error_rep
component of its result.