A FLEXIBLE SENSITIVITY ANALYSIS APPROACH FOR UNMEASURED CONFOUNDING WITH MULTIPLE TREATMENTS AND A BINARY OUTCOME WITH APPLICATION TO SEER-MEDICARE LUNG CANCER DATA

Keyphrases

• Lung Cancer 100%
• Sensitivity Analysis 100%
• Cancer Data 100%
• Unmeasured Confounding 100%
• Multiple Treatments 100%
• Binary Outcome 100%
• SEER-Medicare 100%
• Treatment Effect 40%
• Causal Inference 40%
• Effect Estimates 20%
• Treatment Outcome 20%
• Early-stage Lung Cancer 20%
• Medicare Data 20%
• Causal Effect 20%
• Parameter Sensitivity 20%
• Seamless Integration 20%
• Ment 20%
• Sampling Variability 20%
• Bayesian Framework 20%
• Bayesian Additive Regression Trees 20%
• Randomized Experiment 20%
• Monte Carlo Sensitivity Analysis 20%
• Ignorable 20%
• Ignorability Assumption 20%
• Nested multiple Imputation 20%

Mathematics

• Confounding 100%
• Bayesian 40%
• Causal Inference 40%
• Treatment Effect 40%
• Effect Estimate 20%
• Monte Carlo 20%
• Causal Effect 20%
• Multiple Imputation 20%
• Regression tree 20%
• Ignorability 20%