Efficiently Weighted Estimating Equations with Application to Proportional Excess Hazards
A general approach to estimation, that can lead to efficient estimation in two stages, is presented. The method will not always be available, but sufficient conditions for efficiency are provided together with four examples of its use: (1) estimation of the odds ratio in 1:M matched case-control studies with a dichotomous exposure variable; (2) estimation of the relative hazard in a two-sample survival setting; (3) estimation of the regression parameters in the proportional excess hazards model; and (4) estimation in a partly linear parametric additive hazards model. The method depends upon finding a family of weighted estimating equations, which includes a simple initial equation yielding a consistent estimate and also an equation that yields an efficient estimate, provided the optiomal weights are used.
KeywordsEfficient Score Nuisance Parameter Partial Likelihood Predictable Process Conditional Likelihood
Unable to display preview. Download preview PDF.
- Breslow, N.E., and Day, N.E. (1980), Statistical Methods in Cancer Research: Volume I - The analysis of case-control studies, International Agency for Research on Cancer, Lyon.Google Scholar
- Crowley, J.J., Liu, P.Y., and Voelkel, J.G. (1982), “Estimation of the ratio of hazard functions”, in Crowley, J.J. and Johnson, R.A., editors, Lecture notes — Monograph series 2,Survival Analysis, pp 56–73. Institute of Mathematical Statistics: Hayward, California.Google Scholar
- Sasieni, P. (1995), “Proportional excess hazards,” Submitted.Google Scholar