Moments of Causal Effects
By: Yuta Kawakami, Jin Tian
Potential Business Impact:
Finds how things *really* affect each other.
The moments of random variables are fundamental statistical measures for characterizing the shape of a probability distribution, encompassing metrics such as mean, variance, skewness, and kurtosis. Additionally, the product moments, including covariance and correlation, reveal the relationships between multiple random variables. On the other hand, the primary focus of causal inference is the evaluation of causal effects, which are defined as the difference between two potential outcomes. While traditional causal effect assessment focuses on the average causal effect, this work provides definitions, identification theorems, and bounds for moments and product moments of causal effects to analyze their distribution and relationships. We conduct experiments to illustrate the estimation of the moments of causal effects from finite samples and demonstrate their practical application using a real-world medical dataset.
Similar Papers
Causal Effect Identification in Heterogeneous Environments from Higher-Order Moments
Artificial Intelligence
Finds hidden causes even when data changes.
Causal attribution with confidence
Methodology
Finds if something truly caused an outcome.
Kernel-based estimators for functional causal effects
Methodology
Helps doctors understand how treatments change over time.