Keith Battocchi

If you want to use it for heterogeneity you could still add it to X while changing your model_t to something like a pipeline where the first step is a...

We don't use the `logging` library internally, so I suspect this is due to some transitive dependency.

The answer to this will depend somewhat on what specific models you use. If you use something like CausalForestDML (which will fit a non-parameterized final model) and you also use...

That's correct - `CausalAnalysis` is designed to have a simpler interface so it doesn't expose all of the options that using the DML subclasses directly would provide.

This is an interesting failure mode - if the estimates for E[T|Z,X,W] are always identical to E[T|X,W] then since the final model weights the rows by the estimated variance (E[T|Z,X,W]-E[T|X,W])^2,...

Sorry for the inconvenience; our next release should allow the latest versions of shap and numpy and should avoid this issue.

We've released the beta (v0.15.0b1) for our next version, which should support the latest shap and numpy releases. You can try it out by running `pip install -U --pre econml`

Our DML instances do support (single) non-binary categorical treatments. Your treatment column should have the raw treatment indicators (e.g. this could be something like `[1, 0, 2, 2, 0, 1,...

When there are multiple discrete treatments, we drop the first and the marginal effects should be interpreted as the effect from going from T0 to T1, from T0 to T2,...

Would be great to have more information on exactly what commands you were running when you saw this, as well as what versions of python and pip you're using and...