How to create a function to use in avg_comparisons( transform=)

[daSilva5]

I have a model with a transformed response.

I have tried the following:

  return ((roundlam * x + 1)^(1 / roundlam) - offset)
}

offset<-1

compare_weight<-avg_comparisons(bbiom,variables = list(treatment = "pairwise",transform=function(x) back_boxcox (x, roundlam = p1$roundlam, offset = 1)))
Warning message:
These variables were not found: transform.  Try specifying the `newdata` argument
  explicitly and make sure the missing variable is included. 

I tried to add the newdata, specify the response variable directly, but none of these worked. And could not find examples of how to set up a new back transformation with this package.

Thanks

[vincentarelbundock]

It's hard to help without a fully reproducible MINIMAL working example

But at first glance, I see that you have mismatched parentheses: transform should be on its own, and not inside the variables=list().

Also, probably no space after back_boxcox

[vincentarelbundock]

Also, note that transform will apply as the very last operation. You can do something more complicated by defining your own function in the comparison argument. See the documentation for examples:

https://marginaleffects.com/r/man/comparisons.html

[daSilva5]

Hi,
Removing the space did work! Specifying it in comparison produced the desired result.

back_boxcox <- function(x, roundlam, offset) {
  return ((roundlam * x + 1)^(1 / roundlam) - offset)
}
avg_comparisons(bbiom,comparison = function(hi, lo)mean(back_boxcox(hi,roundlam=p1$roundlam,offset=1))-mean(back_boxcox(lo,roundlam=p1$roundlam,offset=1)))

Cheers,