Closed biometrician closed 1 year ago
I you use print(ABE_object) you will get all the quantities reported in the table by using the same sampling method (as specified in abe.resampling; if you call print.abe.default(ABE_object) abe.resampling will be executed again (once or twice, depending on the original call to abe.resampling) and then the results in the table are reported using appropriate (different) sampling techniques for different quantities reported in the table. I don't know how to change this so that when using print you would actually be using print.abe.default.
Ah, now I understand it.
Georgs and my idea was to use
abe.resampling(type.resampling = c("bootstrap", "mn.bootstrap", "subsampling", "Wallisch2021"))
If one requests "Wallisch2021", then one would get the results according to the recommendations from the Wallisch paper. Otherwise only bootstrap, mn.bootstrap or subsampling is used for all measures. An advantage of this solution would be, that one could easily add another resampling scheme if e.g. recommendations change in the future.
With this solution would it be still necessary to have two different print functions?
This also covers #10, which I close now.
Done, but please double-check everything. Opened another issue regarding the plots.
Thanks Rok! It looks good. :D I will double-check it in the next days.
Currently, to get the results of
ABE_objects
, one has to writeprint.abe.default
. Is it possible to change that, so that one can simply useprint
?If I use
print(ABE_object)
I believe, I get the results of just the first resampled data set. This is very prone to errors.Of course, I understand, that there is also a
print
function for the normal model fits (without resampling).