Propensity Score
AdvancedEnglish: Propensity Score
Probability of treatment assignment given covariates.
Why It Matters
Propensity scores are essential for reducing bias in observational studies, allowing researchers to make more accurate causal inferences. They are widely used in healthcare, social sciences, and economics to evaluate the effectiveness of interventions, ultimately leading to better decision-making and improved outcomes.
Definition
The propensity score is defined as the probability of a unit receiving a particular treatment given a set of observed covariates, mathematically expressed as e(X) = P(T=1 | X), where T is the treatment indicator and X represents the covariates. This score is crucial for addressing selection bias in observational studies by enabling the matching of treated and control units with similar characteristics. The use of propensity scores facilitates the estimation of causal effects through methods such as matching, stratification, and weighting. By balancing covariates between treatment groups, propensity scores help approximate the conditions of a randomized controlled trial, thereby enhancing the validity of causal inferences in causal AI and interpretability.