Targeted Maximum Likelihood Estimator (TMLE) in EconML

[AlxndrMlk]

Hi,

I was recently looking for Targeted Maximum Likelihood Estimator (TMLE) implementations in Python.

To my best understanding EconML does not implement this estimator. Is that correct?

If so, do you think it would be a good idea to add it to EconML?

Existing implementations:

• IBM's CausalLib (notebook, docs)

• zEpid (docs)

[Perhaps there are more]

References:

• Schuler, M. S., & Rose, S. (2017). Targeted Maximum Likelihood Estimation for Causal Inference in Observational Studies. American Journal of Epidemiology, 185(1), 65–73.

• van der Laan, M. & Rubin, D. (2006). Targeted Maximum Likelihood Learning. The International Journal of Biostatistics, 2(1).

[kbattocchi]

Thanks for the suggestion - for the case of a binary outcome this does seem like a potentially valuable addition to the library and it's something we'll consider.

[AlxndrMlk]

Thanks for the reply @kbattocchi

Is there a particular reason behind your suggestion regarding the binary outcome? Do you think that adding continuous outcomes would not be useful in the context of EconML?

Best, Alex