search

py-why / dowhy

DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions. DoWhy is based on a unified language for causal inference, combining causal graphical models and potential outcomes frameworks.
https://www.pywhy.org/dowhy
MIT License
7.03k stars 923 forks source link

Structural Causal Models - Conditional Probabilities, Structural Equations, Gaming Counterfactual Target #1249

Open PMK1991 opened 2 weeks ago

PMK1991 commented 2 weeks ago

For my thesis, I need to use DAGs/SCMs. This is in continuation of my previous bug raised: https://github.com/py-why/dowhy/issues/1241

If I have a GCM, how can I achieve the following:

  1. When I apply an intervention, can I get conditional probabilities of the mutilated graph underneath? The reason I am asking is because, I would like to understand the change in conditional probabilities of the system as we do with Causal Bayesian networks.
  2. Can I get access to the structural equations underneath?
  3. Can I game the system to produce counterfactual target, as we do I Counterfactual Explanations?

Please let me know if these are radical or nonsensical ideas. I am new to SCMs., hence some things may spring from my ignorance.

PMK1991 commented 6 days ago

@bloebp Can you kindly answer this?