1-Why Causal Discovery ?
- Beyond Correlation:Just because two variables are correlated does not mean one causes the other;
- Decision Optimization:Understanding causality helps in designing better policies and interventions;
- Counterfactual Reasoning:Helps answer "what if" scenarios , such as “Would the patient have survived if they had treatment?"
2-Key Concepts in Causal Discovery
- Causal Graphs(Directed Acyclic Graphs-DAGs);
- Causal Markov Condition ;
- Confounders & Colliders ;
3-Methods for Causal Discovery
- Constraint-Based Methods ;
- Score-Based Methods ;
- Functional Causal Models(FCMs) ;
- Granger Causality (For Time Series Data);
- Causal Discovery with Machine Learning;