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
6.88k stars 916 forks source link

Interpreting mean while using logistic regression to estimate causal effect. #1198

Closed athulsudheesh closed 3 weeks ago

athulsudheesh commented 3 weeks ago

How is the mean (which is the reported effect by the package) computed while using logistic regression for estimating causal effect in dowhy? Is the reported effect in cohen's d or odds ratio?

athulsudheesh commented 3 weeks ago

I found the answer myself. The ATE reported by dowhy while using a logistic regression is sometimes called as the risk difference (in social and medical sciences). This article was helpful for me to understand the theory behind the ATE estimation: https://solomonkurz.netlify.app/blog/2023-04-24-causal-inference-with-logistic-regression/ There are formulas one can find in the literature to convert the risk difference to cohen's d (which is the standard measure used in psych sciences).