Reading the evidence
Correlation shows that variables move together under a sample; causation means changing one would change the other under specified conditions. Shared causes, selection and chance can create correlation without causation.
Make it click
Umbrellas and wet roads appear together because rain affects both. Their relationship is real, while changing umbrella count would not stop the roads becoming wet.
Now put that into markets
Market variables share economic drivers and react simultaneously to news. An observed association can still be useful descriptively without establishing a causal mechanism.
Causal inference needs design: experiments, natural experiments, timing, controls or credible assumptions. Regression alone does not supply that design.
Draw plausible causal paths, identify confounders and ask what intervention the claim implies. Separate prediction from explanation.
Controlling for the wrong variable can introduce bias, and temporal order alone is insufficient. Strong causal language needs stronger evidence.
Takeaway
