Neal Fultz
Neal Fultz
It doesn't take levels, but unquoted variable names. maybe join_by or join_on.
IIRC the conversion from matrix to data frame happens at the very end. As a workaround, you could pipe to a subsequent fabricatr call, or use X internally as a...
I think rescaling is generally not what people would expect, eg people don't usually expect sd(rnorm(100)) == 1 exactly - there's sampling variability there, and I think that when ICC...
I don't believe this is unique to potential outcomes. The condition is probably supposed to be: nrow(variable_data) == length(inner_N)
See also https://github.com/HenrikBengtsson/Wishlist-for-R/issues/38
I think you are approaching the limit of understandability for pasting strings together and eval'ing it - I would really recommend spending a little time trying alternative solutions, perhaps the...
See also the code from the shiny app that uses a grid search to cache diagnosis.
Agreed it would better to support a link function - I don't think estimatr has any functionality for that though. On Wed, Aug 15, 2018 at 7:32 AM macartan wrote:...
@lukesonnet @graemeblair - these are testing for precision errors that occur differently on various processor architectures - we should probably just skip the tests if not on an x86.
> The fact that it should be a RHS isn't all that clear, and the fact that it works without the tilde for estimation but not for prediction is problematic....