Abstract:
|
In this work, i will propose estimating equation approaches to estimate the parameters in the semiparametric transformation cure models, which include the semiparametric transformation models for modeling the failure times of the uncured subjects and the logit model for estimating the cure rate. In contrast to the conventional approaches, we only have survival data from the prevalent sampling which is indeed a biased sampling scheme. To account for this sampling bias, more efficient estimating equation approaches will be proposed, including the conditional likelihood approach as the special case. The proposed methods will be examined through the simulation studies and applied to a real data.
|