Causality

How one causes another.

Correlation does not necessarily imply causation. Lack of correlation also does not necessarily imply the lack of causation. A good example can be: “inside temperature not associated with oil burned in the furnace.” or “the sailboat “

Causal relationships are often expressed as a DAG (Directed Acyclic Graph). See also Blom2021conditional for extending it “beyond DAGs”.

Probably one of the most important questions in Causal inference is this: how can we establish causal relationships with observations and experiments? can we perform Causal inference with observational data?

• Rubin causal model
• Causal diagram
• Causal inference
  • Assumptions
    • Ignorability
  • Measures
    • Granger causality
  • Methods
    • Instrumental variables estimation
• Causal language
• External validity

References

• Detecting Causality in Complex Ecosystems
• PNAS causal inference special issue
  • Causal inference in economics and marketing
• Distinguishing Cause from Effect Using Observational Data: Methods and Benchmarks

Memes