It is known that in many missing data models, for example, survival data models, some parameters are root-n estimable while the others are not. When they are, their limiting distributions are often Gaussian and easy to use. When they are not, their limiting di... ...
