Qm
Statistics

Conditioning on a collider opens a fake path

collider-biasbad-controlscausal-inference

Two unrelated traits, skill and luck, both influence whether a fund survives long enough to enter your sample. You study only surviving funds and regress performance on skill, controlling for (or effectively conditioning on) survival.

Explain how conditioning on survival can induce a spurious negative relationship between skill and luck, and why this is a bad control, not a fix.

Your answer

Solving needs a free account

Answers, streaks and solutions unlock when you are signed in. Reading the question and the hint stays free.

Create a free accountSign in

Discussion

Sign in to join the discussion · reading is open to everyone

💡 Discussion rules

  1. No full solutions here. Hints and approaches only.
  2. Complexity, edge cases and intuition are the point.
  3. Interview experiences are welcome. Respect your NDAs.

Loading discussion…

Learn the concepts

The theory behind this question.

Colliders and Collider BiasATE, ATT and Other Treatment Effect EstimandsBad Controls and Post-Treatment BiasBerkson's ParadoxEstimating Conditional Average Treatment Effects With ML

Related questions

The other kind of bias: controlling for a mediatorOver-adjustment: controlling for a proxy of the treatmentWhich of three candidate controls belongs in the regression?Total effect versus direct effect when you control a mediatorEstablishing causation: the index-inclusion effect
All questions →