Conditional Estimation in Two-stage Adaptive Designs

TL;DR

This paper examines various conditional estimation methods in two-stage adaptive designs, comparing their bias and efficiency through analytical and simulation studies, with practical application in clinical trials.

Contribution

It introduces and compares multiple conditional estimators for two-stage adaptive designs, highlighting their properties and practical implementation.

Findings

Rao-Blackwell estimator is conditionally unbiased.

Conditional estimators outperform unconditional ones in bias reduction.

Simulation results demonstrate estimator performance under different scenarios.

Abstract

We consider conditional estimation in two-stage sample size adjustable designs and the following bias. More specifically, we consider a design which permits raising the sample size when interim results look rather promising, and, which keeps the originally planned sample size when results look very promising. The estimation procedures reported comprise the unconditional maximum likelihood, the conditionally unbiased Rao-Blackwell estimator, the conditional median unbiased estimator, and the conditional maximum likelihood with and without bias correction. We compare these estimators based on analytical results and by a simulation study. We show in a real clinical trial setting how they can be applied.