You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
💬 Talk on "Sensitivity Analysis for Inverse Probability Weighting Estimators via the Percentile Bootstrap" (Q. Zhao et al., 2017), for S. Pimentel's "Observational Study Design and Causal Inference" seminar at Berkeley, Spring 2018
A questionnaire containing 40+ questions is given to hundreds of people. People are interviewed about their feelings and hobbies with a goal to find the causal relationship between depression and cognitive impairment, where some questions are related to depression, some to cognitive impairment, and others are confounding. In psychological survey…
Non-parametric variable selection and inference via the outcome-adaptive Random Forest (OARF). Uses the IPTW estimator to estimate the ATE while the propensity score is estimated via OARF. This leads to smaller variance and bias. Only variables that are confounders or predictive of the outcome are selected for the propensity score.
Targeted maximum likelihood estimation (TMLE) enables the integration of machine learning approaches in comparative effectiveness studies. It is a doubly robust method, making use of both the outcome model and propensity score model to generate an unbiased estimate as long as at least one of the models is correctly specified.
The R package trajmsm is based on the paper Marginal Structural Models with Latent Class Growth Analysis of Treatment Trajectories: https://doi.org/10.48550/arXiv.2105.12720.