Technical Reports : Missing Data


Group comparisons involving missing data in clinical trials; a comparison of estimates and power (size) for some simple approaches.

  1. Summary: An investigation of some simple approaches to missing data imputation and how ignoring that imputation could be conservative depending on the model chosen for imputation. in clinical trials, including software implementation.

  2. Ref: Miller ME, Morgan TM, Espeland MA, Emerson SS, Statistics in Medicine 20: 2383-2397 (2001)

A Sensitivity Analysis for Clinical Trials with Informatively Censored Survival Endpoints

  1. Summary: E. Meier’s MS Thesis exploring the robustness of sensitivity analyses based on missing not at random (MNAR) censoring analyzed under a proportional hazards assumption.