🧠🧹 Causality — Mental Hygiene for Data Science

Harness The Power of Why with Causal Tools.

Data does not compensate for assumptions— Judea Pearl

To apply or not to apply, that is the question.

Causal reasoning elevates predictive outcomes by shifting from “what happened” to “what would happen if”. Yet, implementing causality can be challenging or even infeasible in some contexts. This article explores how the very act of assessing its applicability is valuable in its own right since it can improve the scientific rigour of your projects.

The main take away points from this gentle intro to causality are:

• Applicable or not, assessing suitability of causality is good for mental hygiene.

• By applying causal reasoning you will enhance the scientific rigour of your solutions because you will have tools to articulate the problems.

• Causal Models are visual aids for a deeper understanding of mechanisms

• Identifiability is a framework to assess applicability of specific causal questions

This article is targeted at practicing and aspiring data scientists, machine learning engineers, analysts and other practitioners interested in decision-making with causal inference.

No prior knowledge is assumed. For seasoned practitioners I hope to shine light on aspects that may not have been considered 💡.

The content of this article was presented in the recent PyData Global 2024 conference.

For the full article see:

https://bit.ly/causal-mental-hygiene

I will post a YouTube link when available.