There was some discussion on twitter about the fact that the manual
page for as.formula() doesn't mention reformulate(), and indeed the last
example is
## Create a formula for a model with a large number of variables:
xnam <- paste0("x", 1:25)
(fmla <- as.formula(paste("y ~ ", paste(xnam, collapse= "+"))))
which could arguably be better done as
reformulate(xname, response = "y")
I've attached a documentation patch that adds the alternative version
and a \seealso{} link.
Happy to submit to r-bugzilla if requested.
cheers
Ben Bolker
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: reformulate_patch.txt
URL: <https://stat.ethz.ch/pipermail/r-devel/attachments/20220109/f3266495/attachment.txt>
documentation patch for as.formula → reformulate
2 messages · Ben Bolker, Martin Maechler
Ben Bolker on Sun, 9 Jan 2022 16:39:43 -0500 writes:
There was some discussion on twitter about the fact that the manual page for as.formula() doesn't mention reformulate(), and indeed the last example is
## Create a formula for a model with a large number of
variables: xnam <- paste0("x", 1:25) (fmla <-
as.formula(paste("y ~ ", paste(xnam, collapse= "+"))))
which could arguably be better done as
reformulate(xname, response = "y")
I've attached a documentation patch that adds the
alternative version and a \seealso{} link.
Happy to submit to r-bugzilla if requested.
cheers Ben Bolker
x[DELETED ATTACHMENT external: reformulate_patch.txt, plain text] Thanks a lot, Ben! I've committed (+-) it to R-devel as svn rev 81464 Martin