Robustness of Causal Claims

TL;DR

This paper formalizes the concept of robustness in causal claims, providing graphical criteria and algorithms to quantify how resistant causal assertions are to violations of underlying assumptions.

Contribution

It introduces a formal definition of robustness for causal claims, along with graphical conditions and algorithms to measure this robustness.

Findings

Defines a formal measure of robustness for causal claims

Provides graphical criteria to assess robustness

Develops algorithms to compute robustness scores

Abstract

A causal claim is any assertion that invokes causal relationships between variables, for example that a drug has a certain effect on preventing a disease. Causal claims are established through a combination of data and a set of causal assumptions called a causal model. A claim is robust when it is insensitive to violations of some of the causal assumptions embodied in the model. This paper gives a formal definition of this notion of robustness and establishes a graphical condition for quantifying the degree of robustness of a given causal claim. Algorithms for computing the degree of robustness are also presented.