Stefan Wager

That's right; the DR scores used with AIPW are not bounded within the range of the Y. This is simply a consequence of how DR scores are obtained; and doing...

That's a great question. That's a direction that would be natural to explore; however, we haven't (yet) built anything like this in GRF.

If you want parameter choices that are covered by Thm 1 of WA18, I'd recommend just setting parameters directly instead of using the tuning function. (Or, one other option would...

Hi Jinzhao, You can use randomForestCI for classification also, but only on the probability scale. In the the case of a two-class classification problem Y \in {0, 1}, if you...

Here's an example with confidence intervals on new observations: ```R # Make some data... n = 250 p = 100 X = matrix(rnorm(n * p), n, p) Y = rnorm(n)...

Hmm yeah ideally the code would let you specify just one prediction point; that looks like a bug we should fix. In the mean time, yes, appending it to the...

No there's an asymmetry between `x` and `y` in a forest: The forest can only be used to predict `y` from `x`, but not vice-versa (and the same goes for...

This package is deprecated, and no longer under active maintenance. If you want confidence intervals for regression forests, I'd recommend you use the function `regression_forest` in the package `grf`. It...

Sorry for taking so long to get back to you on this; it had fallen off my radar last year. Yes, this definitely seems like a real bug, and needs...

We don't have one currently available. We hope to have one eventually, but our first priority is getting the R implementation to 1.0. If anyone is interested in contributing a...