Erik Sverdrup

I'm not sure about reference, but here is a simple example illustrating the issue with an OOB mean: Let $Y_i \sim Bernoulli(\mu)$, $i=1...n$, with mean $\mu=0.5$. Then $\mu^{(-1)} = \mu...

Hi @corydeburd, That MSE is a reasonable object to consider. Another approach is to treat RDD Forest purely as a data-driven algorithm to find heterogeneous subgroups, i.e: Split data into...

Hi @yusukematsuyama, there's no fixed rule for the train/test, 50/50 and 70/30 are just some common choices. Forests are usually robust wrt tuning parameters, it's hard to say which range...

Hi @minnnjecho, 1. The TOC/RATE can "subsume" the test_calibration exercise, so no plan for future IV support. 2. A significant held-out RATE suggests the tuned forest was able to detect...

Hi @robert702, that's an interesting question. Since the AUTOC can be represented as a weighted ATE ((8) in https://arxiv.org/pdf/2111.07966.pdf) I wonder if RATE + [Crump et al. (2009)'s](https://academic.oup.com/biomet/article-abstract/96/1/187/235329?redirectedFrom=fulltext) subsetting via...

My immediate reaction would be to just do what's posted above, that's one of the reasons I added the `subset` argument to the rate function.

Hi @njawadekar, you can rename `tree$columns` to the names you prefer, then plot.

If you are in a setting with binary W you can have a look at the first expression for Gamma https://grf-labs.github.io/grf/articles/muhats.html to understand what the code above is doing. If...

Hi @n-ihlo, grf doesn't store the value of the criterion it uses to determine splits. Causal forest does CART splits on an appropriate "pseudo-outcome". All the information needed to reconstruct...

Hi @Matt9617, 1. Yes. 2. That's correct. 3. Stefan has a nice video lecture [here](https://youtu.be/fAUmCRgpP6g?list=PLxq_lXOUlvQAoWZEqhRqHNezS30lI49G-&t=1241) on interpreting that kind of calibration exercise. You might find this recent grf feature, [RATE](https://grf-labs.github.io/grf/articles/rate.html),...