Variance Estimation and
Distributional Approximation in Some Semiparametric
Estimation Problems
A long-standing problem in semiparametric survival analysis is the development of reliable inference methods for censored linear regression. To tackle this problem, a number of interesting ideas for variance estimation and, more generally, distributional approximation have been proposed and
implemented. Some of these methods are computationally intensive while others are less so. In this talk, I will give an overview of these methods
and discuss ideas for new approaches, with demonstrations by simulations and examples. Both implementational and theoretical aspects will be
covered.