Propensity Process Sample Clauses

Propensity Process. Estimation and Matching‌ In practice, the Propensity Process must be estimated from the observed data. The challenge for estimating the Propensity Process is that we do not observe the complete covariate process X2(·). We only get to observe X2(·) at a set of discrete time points for each subject. To fix ideas, we first describe how to estimate the GPS using baseline variables X1 and then investigate how to estimate the Propensity Process as well as the GPS using both baseline variables and time-varying variables X2. For assessing the treatment effect, we focus on the approach of matching a treated subject with a subject in the corresponding risk set based on the estimated GPS or Propensity Process, fol- lowed by conducting hypothesis testing using matched pairs. The definition of the risk set is provided in Section 3.3.3.