Causal inference with imperfect instrumental variables

Abstract

Instrumental variables allow for quantification of cause and effect relationships even in the absence of interventions. To achieve this, a number of causal assumptions must be met, the most important of which is the independence assumption, which states that the instrument and any confounding factor must be independent. However, if this independence condition is not met, can we still work with imperfect instrumental variables? Imperfect instruments can manifest themselves by violations of the instrumental inequalities that constrain the set of correlations in the scenario. In this paper, we establish a quantitative relationship between such violations of instrumental inequalities and the minimal amount of measurement dependence required to explain them. As a result, we provide adapted inequalities that are valid in the presence of a relaxed measurement dependence assumption in the instrumental scenario. This allows for the adaptation of existing and new lower bounds on the average causal effect for instrumental scenarios with binary outcomes. Finally, we discuss our findings in the context of quantum mechanics.

The rank of contextuality

“…To confirm the presence of contextuality in actual PDBs, one could test the correlations between attributes and determine if they violate non-contextual inequalities [10,57,58]. This possibility suggests that quantum contextuality methods would detect particular data corruptions analogously to how non-locality techniques serve to detect causal connections in Bayesian networks [59,60]. The preceding suggests research in databases that we intend to develop in future works.…”

The rank of contextuality

“…To confirm the presence of contextuality in actual probabilistic databases, one could test the correlations between attributes and determine if they violate noncontextual inequalities [10,52,53]. This possibility suggests that quantum contextuality methods would detect particular data corruptions analogously to how nonlocality techniques serve to detect causal connections in Bayesian networks [54,55]. The preceding suggests research in databases that we intend to develop in future works.…”