SeGLiR

SeGLiR copied to clipboard

While the Whitehead bias-adjusted estimator seems to work decently, it would be nice to be able to estimate bias conditional on number of samples the test needed to finish.

Furthermore, DeMets and Lan describe modified conditional bias estimators with both low bias and low MSE in their paper "Bias of Point Estimation following a Group Sequential Test", which is applicable to a fully sequential setting as well.

Thank you for your good work, here is the referenced file for safekeeping. tr_157.pdf