Strong consistency properties of nonparametric estimators for randomly censored data,I: The product-limit estimator |
| |
Authors: | A Földes L Rejt? B B Winter |
| |
Institution: | (1) MTA Matematikai Kutató Intézet, Reáltanoda U. 13-15., H-1058 Budapest, Hungary;(2) Department of Mathematics, University of Ottawa, K1N 6N5 Ottawa, Ontario, Canada |
| |
Abstract: | In reliability and survival-time studies one frequently encounters the followingrandom censorship model:X
1,Y
1,X
2,Y
2, is an independent sequence of nonnegative rv's, theX
n'
s having common distributionF and theY
n'
s having common distributionG, Z
n
=min{X
n
,Y
n
},T
n
=IX
n
<-Y
n
]; ifX
n
represents the (potential) time to death of then-th individual in the sample andY
n is his (potential) censoring time thenZ
n
represents the actual observation time andT
n
represents the type of observation (T
n
=O is a censoring,T
n
=1 is a death). One way to estimateF from the observationsZ
1.T
1,Z
2,T
2, (and without recourse to theX
n'
s) is by means of theproduct limit estimator
(Kaplan andMeier 6]). It is shown that
a.s., uniformly on 0,T] ifH(T
–)<1 wherel–H=(l–F) (l–G), uniformly onR if
whereT
F
=sup {x:F(x)<1}; rates of convergence are also established. These results are used in Part II of this study to establish strong consistency of some density and failure rate estimators based on
.The third author's research was partly supported by National Research Council of Canada |
| |
Keywords: | Primary 60F15 62G05 |
本文献已被 SpringerLink 等数据库收录! |
|