In this blog post, we delve into the seven fundamental rules that govern the relationship between causal mechanisms and associations we observe in data. These rules are essential for understanding causal inference, breaking down complex causal graphs into four basic structures: independence, chain, fork, and collider. We explore how causal influence creates correlation between variables, the impact of confounding on correlation, and how random manipulation protects against causal influence. Controlling for variables like confounders or colliders can alter correlations, demonstrating the importance of adjustment in causal inference. However, these rules depend on key assumptions such as consistency and positivity, which are crucial for valid causal inferences.
https://pedermisager.org/blog/seven_basic_rules_for_causal_inference/