EasyDeepLearn

When is the frontdoor criterion useful?

hard

Answer

  • When unmeasured confounders make backdoor impossible, but a fully-mediating variable M exists.
  • E.g.
  • T → M → Y where all T-Y effect goes through M and no confounder acts on M.
  • Frontdoor: P(Y    do(T))  =  ΣmP(Y\; \mid \;\mathrm{do}(T))\; = \;{\Sigma}_{m} P(m    T)    ΣtP(m\; \mid \;T)\; \cdot \;{\Sigma}_{t}' P(Y    m,  t)P(Y\; \mid \;m, \;t) P(t').
  • Classic example: smoking → tar → cancer, if we could measure tar but not the unobserved 'smoking-gene' confounder.
  • Rarely applicable in practice but a landmark result in causal inference theory.
Check yourself — multiple choice
  • Random
  • Uses fully-mediating M: P(Ydo(T))  =  ΣP(Y \mid \mathrm{do}(T))\; = \;{\Sigma} P(mT)P(m \mid T) Σ P(Ym,t)P(Y \mid m, t) P(t'); works even with unmeasured confounders on T; rare in practice
  • Same as backdoor
  • Not real

Frontdoor: mediating M identifies do(T) despite unmeasured T-Y confounders.

#causal-inference

Practise Statistics Fundamentals

215 interview questions in this topic.

Related questions