Asymptotic Theory in Rank Estimation for AFT Model Under Fixed Censorship
Rank based parameter estimation in the accelerated failure time model with right-censoring has been studied rigorously by many authors [(1990), (1991) and (1993)]. A key assumption in establishing the asymptotic normality for the rank estimator is that, conditional on covariates, the censoring time has a density function. This assumption may be violated in many situations. It may also be non-intrinsic in that the limiting variance formula does not involve such a density. We show in this paper that the usual asymptotic theory for rank estimation can indeed be established without this stringent and unnatural assumption.
KeywordsAccelerated failure time model asymptotic normality exponential inequality fixed censoring linear rank statistics survival data
Unable to display preview. Download preview PDF.
- 1.Aalen, O. O. (1975). Statistical inference for a family of counting processes, Ph.D. thesis, University of California, Berkeley.Google Scholar
- 3.Cox, D. R. and Oakes, D. (1984). Analysis of Survival Data, Chapman & Hall, London, England.Google Scholar
- 14.Mantel, N. and Haenszel, W. (1959). Statistical aspects of the analysis of data from retrospective studies of disease, Journal of National Cancer Institute, 22, 719–748.Google Scholar
- 19.Pollard, D. (1990). Empirical Processes: Theory and Application. NSFCBMS regional conference series in Probability and Statistics, Vol 2, Institute of Mathematical Statistics and American Statistical Association, Hayward, California.Google Scholar