Disproving Causal Relationships Using Observational Data
提出一种利用观测数据检验因果假设的方法，通过引入第三个变量来否定两个变量间的因果关系，并区分强否定与弱否定，蒙特卡洛模拟显示强否定在小样本下也高度可靠。
Abstract Economic theory is replete with causal hypotheses that are scarcely tested because economists are generally constrained to work with observational data. We describe a method for testing a hypothesis that one observed random variable causes another. Contingent on a sufficiently strong correspondence between the two variables, an appropriately related third variable can be employed for the test. The logic of the procedure naturally suggests strong and weak grounds for rejecting the causal hypothesis. Monte Carlo results suggest that weakly grounded rejections are unreliable for small samples, but reasonably reliable for large samples. Strongly grounded rejections are highly reliable, even for small samples.