I am confused, you _can_ measure genotype, and genotype here can be argued as a valid confounding…

Arindam Basu

·Nov 18, 2020

I am confused, you _can_ measure genotype, and genotype here can be argued as a valid confounding variable, and hence there is backdoor path? Can you please explain if I am getting it wrong?

…irect effect of Smoking (X) on Cancer (Y), while being unable to directly measure the Genotype (U). From the DAG we can see that no variable satisfies the back-door criterion as U is unmeasured, so we can immediately write:

Causal Inference — Part XII — Front-door Criterion

Bruno Gonçalves