Adding test planning capabilities into CausalPy

[cetagostini]

Hello team,

As we discussed, I'm creating this issue to talk about a new feature that can significantly improve the utility of causalpy for experimenters. The idea it's to bring capabilities which help to do planning previous the experiment.

From the top of my head, I have a few ideas:

• Power analysis by ROPE (Determinate the false positive/negative rate prior intervention).
• Automatic covariates selection for any time series method (e.g: ITS, SCG).
• Integration of "posterior tail area probability".
• Automatic selection of experiment length (Should we do 10 days or 30 days experiments).

All of this are separate problems, which probably we should take on several PRs. But I would love to hear your ideas.

@drbenvincent @JakePiekarski314

[cetagostini]

I could take again number one: #276 solving only ROPE power analysis :) So, users can run regression before intervention, and get get some information.