Pinned post
WP๐จ with @danzeltzer @DanielNevo @B_WNSN & Walk.
If you used an IV, you likely ran a falsification or placebo test for exogeneity. We show these tests can be *too harsh*. They can flag issues even if the IV is exogenous.#EconTwitter #CausalInference
arxiv.org/abs/2312.15624
๐งต>>
If you used an IV, you likely ran a falsification or placebo test for exogeneity. We show these tests can be *too harsh*. They can flag issues even if the IV is exogenous.#EconTwitter #CausalInference
arxiv.org/abs/2312.15624
๐งต>>