The Evaluation of Causal Effects in Studies with an Unobserved Exposure/Outcome Variable: Bounds and Identification
This work addresses challenges in causal inference for practical fields where observing key variables is difficult or expensive, offering incremental improvements in bounding and identification methods.
The paper tackles the problem of evaluating causal effects from observational data when an exposure or outcome variable is unobserved, proposing identifiability criteria for multi-category unobserved variables and providing tight bounds for cases with unmeasured variables between unobserved outcomes and proxies.
This paper deals with the problem of evaluating the causal effect using observational data in the presence of an unobserved exposure/ outcome variable, when cause-effect relationships between variables can be described as a directed acyclic graph and the corresponding recursive factorization of a joint distribution. First, we propose identifiability criteria for causal effects when an unobserved exposure/outcome variable is considered to contain more than two categories. Next, when unmeasured variables exist between an unobserved outcome variable and its proxy variables, we provide the tightest bounds based on the potential outcome approach. The results of this paper are helpful to evaluate causal effects in the case where it is difficult or expensive to observe an exposure/ outcome variable in many practical fields.