基于Copula函数的相依删失数据的非参数统计推断
|
摘要
本文针对生存分析中相依区间删失两种类型数据,在Copula理论框架下,讨论失效时间分布函数的非参数统计推断问题.首先,给出了相依I型区间删失数据（即现状数据）基于Copula函数框架下的两种非参数估计.分别给出了基于阿基米德Copula函数下,失效时间分布函数的非参数估计的完整解析表达式,也给出了在一般Copula函数下的估计方程.并证明了非参数估计的唯一性.通过模拟分析给出了很好的估计结果.并应用于肿瘤实验数据中.其次,给出了相依II型区间删失数据下,失效时间的分布函数的非参数估计.对于区间删失中的相依问题,我们采用Copula模型建立失效时间变量和删失变量之间的相依性,从而构造估计方程得到失效时间的分布函数的非参数估计.通过模拟研究得到给出估计的有效性,并应用于接受血液因子受感染的血友病患者数据的研究中.
Statistical analysis problem of analyzing all kinks of data about lifetime, survivaltime or failure time arises in a number of applied felds,such as engineering, biomedical,public health, epidemiology,economics and demography.From product durability studyto all kinks of human’s disease study, we need survival analysis’s application. Censoreddata arises when an individual’s life length is known to occur only in a certain periodof time. Possible censoring schemes are right censoring, where all that is known is thatthe individual is still alive at a given time, left censoring when all that is known isthat the individual has experienced the event of interest prior to the start of the study,or interval censoring, where the only information is that the event occurs within someinterval. Truncation schemes are left truncation, where only individuals who survive asufcient time are included in the sample and right truncation, where only individualswho have experienced the event by a specifed time are included in the sample. Theissues of censoring and truncation are dealt more carefully. A common feature of thesedata sets is they contain either censored or truncated observations. Interval censoringis a type of censoring that has become increasingly common in the areas that producefailure time data. In the past20years or so, a voluminous literature on the statisticalanalysis of interval-censored failure time data has appeared. Interval censored data aredivided into two types, one is current status data, that is case I interval-censored data;another interval censored data is Case II data. If the failure time and the observationtime can be assumed to be independent, several methods have been developed forthe problem. But in practice are often faced with the failure time variable T andthe observation time variable C or U, V does not meet the independence assumption,we call dependent censoring. Copula theory in modeling the dependence betweenvariables play an increasingly large role. Here we will focus on the situation where theindependent assumption does not hold for two kinds of interval censoring data typeand propose estimation procedures for the distribution of failure time data under the Copula model framework and apply these methods to tumorigenicity experiments andAIDS data. Here we introduce the main results of this paper.
First, Chapter2discusses nonparametric estimation of a survival function whenone observes only current status data (McKeown and Jewell,2010; Sun,2006; Sunand Sun,2005). In this case, each subject is observed only once and the failure timeof interest is observed to be either smaller or larger than the observation or censoringtime. If the failure time and the observation time can be assumed to be independent,several methods have been developed for the estimation. Here we will focus on thesituation where the independent assumption does not hold, which often occurs in,for example, tumorigenicity experiments. For the problem, two simple estimationprocedures are proposed under the Copula model framework. The estimates allowone to perform sensitivity analysis or identify the shape of a survival function amongother uses. A simulation study performed indicate that the two methods work well andthey are applied to a motivating example from a tumorigenicity study. Nonparametricestimation of a survival function is often the frst task in performing the failure timedata analysis and many procedures have been developed for the problem (Kalbfeisch&Prentice,2002; Hu&Lawless,1996; Hu et al.,1998; Klein&Moeschberger,2002).However, most of the existing procedures are for right-censored failure time data. Thispaper discusses nonparametric estimation of a survival function when one observesonly current status data (McKeown&Jewell,2010; Sun,2006; Zhang et al.,2005).In this situation, each subject is observed only once and the failure time of interest isobserved to be either smaller or larger than the observation time. Several procedureshave been developed for the estimation problem in the literature if the failure time andthe observation time can be assumed to be independent. In this paper, we will focus onthe situation where the independent assumption does not hold, for which there seemsno nonparametric estimation procedure available.
This study was motivated by the analysis of tumorigenicity experiments, in whichcurrent status data routinely occur and the estimation of tumor prevalence function isoften required (Keiding,1991). This is because that in this situation, the failure timeof interest is usually the time to tumor onset, which is commonly not observed. Insteadonly the death or sacrifce time of an animal, serving as the observation time here, isknown (Hoel&Walburg,1972; Lagakos&Louis,1988). If the tumor is nonlethal, thenit is usually reasonable to assume that the time to tumor onset and the death time areindependent. Of course, if the tumor is lethal, the estimation is straightforward as wewould have right-censored data on the time to tumor onset. On the other hand, it iswell-known that most types of tumors are between lethal and nonlethal, meaning that the time to tumor onset and the death time are related. Lagakos and Louis (1988)discussed several examples of such data and pointed out that for the problem, thesensitivity type analysis has to be performed as the correlation between the two timevariables, often referred to as lethality, cannot be estimated. The proposed estimateswould make the sensitivity analysis possible. Another example will be discussed below.
As mentioned above, several procedures have been proposed for nonparametricestimation of the survival or cumulative distribution function based on current statusdata when the failure time of interest and the observation time can be assumed to beindependent (Sun,2006). In the following, such data will be referred to as independentcurrent status data and otherwise as dependent current status data. For example,in this case, the maximum likelihood estimate of the survival function can be easilyderived by using the pool-adjacent-violator algorithm for the isotonic regression. Alsofor independent current status data, many authors have considered other analysis is-sues on them such as treatment comparison and regression analysis (Sun,2006). Fordependent current status data, on the other hand, there exists only limited literaturein general except in the feld of tumorigenicity experiments. Among others, Zhang etal.(2005) considered regression analysis of general dependent current status data andproposed an estimating equation-based inference procedure. A large literature existsfor tumorigenicity experiments and in this case, a common approach is to apply thethree-state model to describe the tumor process. However, with respect to estimationof tumor growth and prevalence function, most of the existing methods are parametricprocedures or assume that tumors are lethal or nonlethal. It is obvious that a non-parametric estimate would be very useful for identifying the shape of the prevalencefunction or model checking among other uses.
Several approaches are commonly used to deal with correlated failure times orthe failure time data with dependent censoring (Hougaard,1986). One is the frailtyapproach that models the correlation by using some latent variables and another isthe Copula model approach. Let T and X denote the two possibly related randomvariables, F and G their marginal distributions, respectively, and H their joint distri-bution. Then it has been shown (Nelsen,2006) that there exists a Copula functionC(u, v), defned on I2with C(u,0)=C(0, v)=0, C(u,1)=u and C(1, v)=v,such thatH(t, x)=C(F (t), G(x)), t≥0, x≥0.(1)
If F and G are continuous, C is unique. Furthermore, for any given Copula functionC and marginal distribution functions F and G, the function H defned in equation(1) is the corresponding distribution function. Amonth others, Zheng and Klein (1995) employed the Copula model approach for estimation of a survival function in the case of dependent right-censored data. In the following, we will adopt the same approach. However, it should be noted that the two situations are quite different as in the latter, the data structure is much more complex and the observed relevant information is extremely less.
In the following, we will assume that T and X represent the failure time of interest and the observation time, respectively. The main goal of this paper is to discuss nonparametric estimation of F under model equation (1). In Section2, we will show that if the Copula function C is known, the survival function F of interest can be uniquely identified. In Section3, we will present two simple consistent estimates of F and both can be easily obtained. The first one is developed for the Archimedean Copula functions and has a closed form, while the second one is for general Copula functions. Section4gives some numerical results and in Section5, we apply the methods to a set of current status data arising from a tumorigenicity experiment. Section6concludes with some discussion and remarks. In the following, we will assume that both T and X are continuous variables.
Consider a survival study that involves n independent subjects and gives the observed data{Xi,δi=I(Xi≥Ti); i=1,...,n}, the i.i.d. replications of{X,δ I(X≥T)}. Let F, G and H be defined as before and suppose that P(Ti=Xi)=0. It is easy to see that given the observed current status data, one can directly estimate the following three functions P1(x)=P(X≤x)=G(x), p2(x)=P(X> x, T Let C be the Copula function defined in (1). Then we have C(u,v)=H(F-1(u),G-1(v)), u,v∈[0,1]. Let μc denote the probability measure corresponding to C. Then a simple calculation gives p2(x)=μC(Ax), where Ax={(u,v):0≤u≤FG-1(v), G(x)≤v<≤1. The following theorem establishes the identifiability of F for the situation considered here.
Theorem1Suppose the marginal distribution functions F and G are continuous and strictly increasing over (0,∞). Also suppose μc(E)>0for any open set E in [0,1]×[0,1].Then given the Copula function C, the marginal distribution function F is uniquely determined by p1(x) and p2(x) or p1(x) and p3(x).In this subsection, we will assume that C is an Archimedean Copula function given by Cφ(u,v)=φ-1[φ(u)+φ(v)],φ∈φ,(2) where φ denotes the class of functions φ:[0,1]→[0,∞] with continuous first and second derivatives and satisfying φ(1)=0,φ'<0,φ">0,0 Theorem2Suppose that the conditions in Theorem1hold and also suppose that φ(t)→∞and|φ'(t)|→∞when t→0.Then F can be expressed as f(t)=φ-1{φ(φ')-1(g(t)φ'(G(t))/D1t))-φ(G(t))}(3) where D1(t)=dP(T we consider estimation of the marginal function F for general Copula functions. Let G, ps and g be denned as before. By Theorem1, F can be uniquely determined by G and p3given C. So it is natural to develop an estimate of F by using G and p3.
Let0 Secondly, Chapter3gives the simple estimation procedures based the Copula in interval censoring data with informative censoring. Nonparametric estimation of a sur-vival function is one of the most commonly asked questions in the analysis of failure time data and for this, a number of procedures have been developed under various types of censoring structures (Kalbfleisch and Prentice,2002). In particular, several algorithms are available for interval-censored failure time data with independent cen-soring mechanism (Sun,2006; Turnbull,1976). In this paper, we consider the interval-censored data where the censoring mechanism may be related to the failure time of interest, for which there does not seem to exist a nonparametric estimation procedure. It is well-known that with informative censoring, the estimation is possible only under some assumptions. To attack the problem, we take a Copula model approach to model the relationship between the failure time of interest and censoring variables and present a simple nonparametric estimation procedure. The method allows one to conduct a sensitivity analysis among others.
Statistical analysis of interval-censored failure time data has recently attracted a great deal of attention (Finkelstein,1986; Sun,2006; Zhang and Sun,2010). By interval-censored data, we usually mean the failure time data in which the failure time of interest is observed to belong to some intervals instead of to be exactly known. Such data often occur in many fields including clinical trials and longitudinal studies. One common example occurs in medical or health studies that entail periodic follow-up. In this situation, an individual due for the pre-scheduled observations for a clinically observable change in disease or health status may miss some observations and return with a changed status. Accordingly, we only know that the true event time is greater than the last observation time at which the change has not occurred and less than or equal to the first observation time at which the change has been observed to occur, thus giving an interval which contains the real (but unobserved) time of occurrence of the change. In this paper, we will discuss nonparametric estimation of a survival function in these situations. Furthermore, the censoring mechanism may be related to the failure time of interest.
A more specific and well-known example of interval-censored data was discussed in Finkelstein (1986) among others. The data arose from a retrospective study on early breast cancer patients who had been treated at the Joint Center for RadiationTherapy in Boston between1976and1980. During the study, the patients were giveneither radiation therapy alone or radiation therapy plus adjuvant chemotherapy andsupposed to be seen at clinic visits every4to6months. However, actual visit timesdifer from patient to patient, and times between visits also vary. At visits, physiciansevaluated the cosmetic appearance of the patient such as breast retraction, a responsethat has a negative impact on overall cosmetic appearance. With respect to the timeto breast retraction, some patients did not experience breast retraction during thestudy, thus giving right-censored observations. For the other patients, the observationsare intervals given by the last clinic visit time at which breast retraction had not yetoccurred and the frst clinic visit time at which breast retraction was detected. Thatis, only interval-censored data were observed for the time to breast retraction. Anotherexample will be discussed below.
Nonparametric estimation of a survival function is often the frst task in perform-ing the failure time data analysis and many procedures have been developed for theproblem (Kalbfeisch&Prentice,2002; Hu&Lawless,1996; Hu et al.,1998; Klein&Moeschberger,2003). However, most of the existing procedures are for right-censoredfailure time data. Several procedures are also avaliable for interval-censored data (Gen-tlemand and Geyer,1994; Sun,2006; Turnbull,1976; Wellner and Zhan,1997). Forexample, the simplest procedure is perhaps the self-consistency algorithm given byTurnbull (1976). However, all of the available procedures are for the situation of inde-pendent censoring mechanism. It does not seem to exist a nonparametric estimationprocedure for the situation where the censoring mechanism may be related to the failuretime of interest or informative. It is well-known that with the presence of informativecensoring, the survival function may not be identifable unless under some assumption-s about the relationship between the failure time of interest and censoring variables(Zheng and Klein,1995). Nevertheless, a nonparametric estimate would be very usefulas it would allow one to, for example, conduct some sensitivity analysis or identify theshape of the underlying survival function.
Two approaches are commonly used to deal with failure time data with dependentcensoring (Hougaard,1986). One is the frailty approach that models the relationshipby using some latent variables (Zhang et al.,2005) and the other is the Copula modelapproach. In this paper, we will take the Copula model approach, which will bebriefy described in Section3.1along some notation. In Section3.2, we will presenta nonparametric estimation procedure for interval-censored data where the censoringvariables may be related to the failure time variable of interest. The procedure can be easily implemented and the key idea behind it is to divide the observed data into two sets of current status data, for which the estimation is relatively easy. Section3.3gives some numerical results and in particular, the method is illustrated by a set of interval-censored data arising from an AIDS study.
In this chapter, we will define the data structure and describe some notation and assumptions that will be used throughout the paper. In particular, the Copula model will be briefly discussed.
Consider a failure time study and let T denote the failure time of interest. Suppose that instead of observed exactly, the observation on T is characterized by two random variables U and V with U As mentioned above, the Copula model provides a very flexible approach to model the relationship among correlated random variables (Hougaard,1986; Nelsen,2006). Define I=[0,1]. A k-dimensional Copula is a function C from Ik to I such that for all (u1,...,uk) E Ik, C(ui,...,uk)=0if at least one coordinate is0and if all coordinates are1except uj, then C(u1,...,Uk)=Uj. In particular, if C is a two-dimensional Copula function, we have that C(u,0)=C(0,v)=0, C(u,1)=u, C(1,v)=v for all u,v∈I2, and C(u2,v2)-C(u2,v1)-C(u1,v2)+C(u1,v1)≥0, for all u1, u2, v1, v2in I such that u1 One of the important properties about the Copula model is given by the Sklar theorem. It states that if H is an k-dimensional distribution function with the marginal distributions G1,..., Gk, then there exists an k-dimensional Copula function C such that H(x1,...,xk)=C(Gi(χ1),...,Gk(xk)),(6) for all (χ1,...,χk). Furthermore, if G1,...,Gk are continuous, then C is unique. On the other hand, for any given k-dimensional Copula function C and k one-dimensional distribution funtions G1,...,Gk,the function H defined above is a distribution function with the marginal distributions G1,…,Gk. It is easy to see that a key feature of the above expression is that the Copula function C defines or characterizes the correlation or the relationship among the k concerned random variables.
Another advantage of the Copula model approach is its flexibility as there exist many different Copula functions.Among them,one class of the Copula functions that are commonly used is the Archimedean Copula functions with the three-dimensional function defined as C3(u1,u2,u3)=φ-1(φ(u1)+φ(u2)+φ(u3)),φ∈Φ,(7) whereΦdenotes the class of functionsφ:Ⅰ→[0,∞]with continuous first and second derivatives and satisfying φ(1)=0,φ'<0,φ">0,0 In the following,we will assume that the joint distribution function H of T,U and W can be described by a Copula function as in(6)and discuss the estimation of the marginal distribution function F1of T.The similar idea was used by,among others Zheng and Klein(1995)for estimation of a survival function based on dependent right-censored time data data.It should be noted that the data structure considered here is much more complex and also the observed relevant information here is extremely less. Now we will consider estimation of the marginal distribution function F1.Suppose that the observed data are n i.i.d.replications of{U,V,δ1=I(T≤U),δ2=I(U To present the estimation procedure,let C denote the Copula function defined in(6)for the joint distribution function H of T,U and V and C(1,2)(u1,u2)and C(1,3)(u1,u3)the resulting marginal joint distribution functions of(T,U)and(T,V), respectively.Define p1(x)=P(T≤U,U≤x) and P2(y)=P(T≤V,V≤y). Then one can show that and where C2(1,2)(u1,u2)=ac(1,2)(u1,u2)/au1and C3(1,3)=ac(1,3)(u1,u3)/au3.As mentioned before,one can easily estimate F2and F3by their empirical estimates since we have complete data on the Ui's and Vi's. This is actually also true for p1(x)and p2(x)and thus it is natural to estimate F1based on(8)and(9)by replacing F2,F3, p1(x)and p2(x)with their empirical estimates.
Specifically,let x0=0 Note that the estimate F1given above is defined only at the xj's. For its value between the xj's,it is natural to define them through the linear extrapolation.It can be easily shown that the estimate proposed above is consistent and nondecreasing for large n.
Statistical analysis problem of analyzing all kinks of data about lifetime, survivaltime or failure time arises in a number of applied felds,such as engineering, biomedical,public health, epidemiology,economics and demography.From product durability studyto all kinks of human’s disease study, we need survival analysis’s application. Censoreddata arises when an individual’s life length is known to occur only in a certain periodof time. Possible censoring schemes are right censoring, where all that is known is thatthe individual is still alive at a given time, left censoring when all that is known isthat the individual has experienced the event of interest prior to the start of the study,or interval censoring, where the only information is that the event occurs within someinterval. Truncation schemes are left truncation, where only individuals who survive asufcient time are included in the sample and right truncation, where only individualswho have experienced the event by a specifed time are included in the sample. Theissues of censoring and truncation are dealt more carefully. A common feature of thesedata sets is they contain either censored or truncated observations. Interval censoringis a type of censoring that has become increasingly common in the areas that producefailure time data. In the past20years or so, a voluminous literature on the statisticalanalysis of interval-censored failure time data has appeared. Interval censored data aredivided into two types, one is current status data, that is case I interval-censored data;another interval censored data is Case II data. If the failure time and the observationtime can be assumed to be independent, several methods have been developed forthe problem. But in practice are often faced with the failure time variable T andthe observation time variable C or U, V does not meet the independence assumption,we call dependent censoring. Copula theory in modeling the dependence betweenvariables play an increasingly large role. Here we will focus on the situation where theindependent assumption does not hold for two kinds of interval censoring data typeand propose estimation procedures for the distribution of failure time data under the Copula model framework and apply these methods to tumorigenicity experiments andAIDS data. Here we introduce the main results of this paper.
First, Chapter2discusses nonparametric estimation of a survival function whenone observes only current status data (McKeown and Jewell,2010; Sun,2006; Sunand Sun,2005). In this case, each subject is observed only once and the failure timeof interest is observed to be either smaller or larger than the observation or censoringtime. If the failure time and the observation time can be assumed to be independent,several methods have been developed for the estimation. Here we will focus on thesituation where the independent assumption does not hold, which often occurs in,for example, tumorigenicity experiments. For the problem, two simple estimationprocedures are proposed under the Copula model framework. The estimates allowone to perform sensitivity analysis or identify the shape of a survival function amongother uses. A simulation study performed indicate that the two methods work well andthey are applied to a motivating example from a tumorigenicity study. Nonparametricestimation of a survival function is often the frst task in performing the failure timedata analysis and many procedures have been developed for the problem (Kalbfeisch&Prentice,2002; Hu&Lawless,1996; Hu et al.,1998; Klein&Moeschberger,2002).However, most of the existing procedures are for right-censored failure time data. Thispaper discusses nonparametric estimation of a survival function when one observesonly current status data (McKeown&Jewell,2010; Sun,2006; Zhang et al.,2005).In this situation, each subject is observed only once and the failure time of interest isobserved to be either smaller or larger than the observation time. Several procedureshave been developed for the estimation problem in the literature if the failure time andthe observation time can be assumed to be independent. In this paper, we will focus onthe situation where the independent assumption does not hold, for which there seemsno nonparametric estimation procedure available.
This study was motivated by the analysis of tumorigenicity experiments, in whichcurrent status data routinely occur and the estimation of tumor prevalence function isoften required (Keiding,1991). This is because that in this situation, the failure timeof interest is usually the time to tumor onset, which is commonly not observed. Insteadonly the death or sacrifce time of an animal, serving as the observation time here, isknown (Hoel&Walburg,1972; Lagakos&Louis,1988). If the tumor is nonlethal, thenit is usually reasonable to assume that the time to tumor onset and the death time areindependent. Of course, if the tumor is lethal, the estimation is straightforward as wewould have right-censored data on the time to tumor onset. On the other hand, it iswell-known that most types of tumors are between lethal and nonlethal, meaning that the time to tumor onset and the death time are related. Lagakos and Louis (1988)discussed several examples of such data and pointed out that for the problem, thesensitivity type analysis has to be performed as the correlation between the two timevariables, often referred to as lethality, cannot be estimated. The proposed estimateswould make the sensitivity analysis possible. Another example will be discussed below.
As mentioned above, several procedures have been proposed for nonparametricestimation of the survival or cumulative distribution function based on current statusdata when the failure time of interest and the observation time can be assumed to beindependent (Sun,2006). In the following, such data will be referred to as independentcurrent status data and otherwise as dependent current status data. For example,in this case, the maximum likelihood estimate of the survival function can be easilyderived by using the pool-adjacent-violator algorithm for the isotonic regression. Alsofor independent current status data, many authors have considered other analysis is-sues on them such as treatment comparison and regression analysis (Sun,2006). Fordependent current status data, on the other hand, there exists only limited literaturein general except in the feld of tumorigenicity experiments. Among others, Zhang etal.(2005) considered regression analysis of general dependent current status data andproposed an estimating equation-based inference procedure. A large literature existsfor tumorigenicity experiments and in this case, a common approach is to apply thethree-state model to describe the tumor process. However, with respect to estimationof tumor growth and prevalence function, most of the existing methods are parametricprocedures or assume that tumors are lethal or nonlethal. It is obvious that a non-parametric estimate would be very useful for identifying the shape of the prevalencefunction or model checking among other uses.
Several approaches are commonly used to deal with correlated failure times orthe failure time data with dependent censoring (Hougaard,1986). One is the frailtyapproach that models the correlation by using some latent variables and another isthe Copula model approach. Let T and X denote the two possibly related randomvariables, F and G their marginal distributions, respectively, and H their joint distri-bution. Then it has been shown (Nelsen,2006) that there exists a Copula functionC(u, v), defned on I2with C(u,0)=C(0, v)=0, C(u,1)=u and C(1, v)=v,such thatH(t, x)=C(F (t), G(x)), t≥0, x≥0.(1)
If F and G are continuous, C is unique. Furthermore, for any given Copula functionC and marginal distribution functions F and G, the function H defned in equation(1) is the corresponding distribution function. Amonth others, Zheng and Klein (1995) employed the Copula model approach for estimation of a survival function in the case of dependent right-censored data. In the following, we will adopt the same approach. However, it should be noted that the two situations are quite different as in the latter, the data structure is much more complex and the observed relevant information is extremely less.
In the following, we will assume that T and X represent the failure time of interest and the observation time, respectively. The main goal of this paper is to discuss nonparametric estimation of F under model equation (1). In Section2, we will show that if the Copula function C is known, the survival function F of interest can be uniquely identified. In Section3, we will present two simple consistent estimates of F and both can be easily obtained. The first one is developed for the Archimedean Copula functions and has a closed form, while the second one is for general Copula functions. Section4gives some numerical results and in Section5, we apply the methods to a set of current status data arising from a tumorigenicity experiment. Section6concludes with some discussion and remarks. In the following, we will assume that both T and X are continuous variables.
Consider a survival study that involves n independent subjects and gives the observed data{Xi,δi=I(Xi≥Ti); i=1,...,n}, the i.i.d. replications of{X,δ I(X≥T)}. Let F, G and H be defined as before and suppose that P(Ti=Xi)=0. It is easy to see that given the observed current status data, one can directly estimate the following three functions P1(x)=P(X≤x)=G(x), p2(x)=P(X> x, T
Theorem1Suppose the marginal distribution functions F and G are continuous and strictly increasing over (0,∞). Also suppose μc(E)>0for any open set E in [0,1]×[0,1].Then given the Copula function C, the marginal distribution function F is uniquely determined by p1(x) and p2(x) or p1(x) and p3(x).In this subsection, we will assume that C is an Archimedean Copula function given by Cφ(u,v)=φ-1[φ(u)+φ(v)],φ∈φ,(2) where φ denotes the class of functions φ:[0,1]→[0,∞] with continuous first and second derivatives and satisfying φ(1)=0,φ'<0,φ">0,0
Let0
Statistical analysis of interval-censored failure time data has recently attracted a great deal of attention (Finkelstein,1986; Sun,2006; Zhang and Sun,2010). By interval-censored data, we usually mean the failure time data in which the failure time of interest is observed to belong to some intervals instead of to be exactly known. Such data often occur in many fields including clinical trials and longitudinal studies. One common example occurs in medical or health studies that entail periodic follow-up. In this situation, an individual due for the pre-scheduled observations for a clinically observable change in disease or health status may miss some observations and return with a changed status. Accordingly, we only know that the true event time is greater than the last observation time at which the change has not occurred and less than or equal to the first observation time at which the change has been observed to occur, thus giving an interval which contains the real (but unobserved) time of occurrence of the change. In this paper, we will discuss nonparametric estimation of a survival function in these situations. Furthermore, the censoring mechanism may be related to the failure time of interest.
A more specific and well-known example of interval-censored data was discussed in Finkelstein (1986) among others. The data arose from a retrospective study on early breast cancer patients who had been treated at the Joint Center for RadiationTherapy in Boston between1976and1980. During the study, the patients were giveneither radiation therapy alone or radiation therapy plus adjuvant chemotherapy andsupposed to be seen at clinic visits every4to6months. However, actual visit timesdifer from patient to patient, and times between visits also vary. At visits, physiciansevaluated the cosmetic appearance of the patient such as breast retraction, a responsethat has a negative impact on overall cosmetic appearance. With respect to the timeto breast retraction, some patients did not experience breast retraction during thestudy, thus giving right-censored observations. For the other patients, the observationsare intervals given by the last clinic visit time at which breast retraction had not yetoccurred and the frst clinic visit time at which breast retraction was detected. Thatis, only interval-censored data were observed for the time to breast retraction. Anotherexample will be discussed below.
Nonparametric estimation of a survival function is often the frst task in perform-ing the failure time data analysis and many procedures have been developed for theproblem (Kalbfeisch&Prentice,2002; Hu&Lawless,1996; Hu et al.,1998; Klein&Moeschberger,2003). However, most of the existing procedures are for right-censoredfailure time data. Several procedures are also avaliable for interval-censored data (Gen-tlemand and Geyer,1994; Sun,2006; Turnbull,1976; Wellner and Zhan,1997). Forexample, the simplest procedure is perhaps the self-consistency algorithm given byTurnbull (1976). However, all of the available procedures are for the situation of inde-pendent censoring mechanism. It does not seem to exist a nonparametric estimationprocedure for the situation where the censoring mechanism may be related to the failuretime of interest or informative. It is well-known that with the presence of informativecensoring, the survival function may not be identifable unless under some assumption-s about the relationship between the failure time of interest and censoring variables(Zheng and Klein,1995). Nevertheless, a nonparametric estimate would be very usefulas it would allow one to, for example, conduct some sensitivity analysis or identify theshape of the underlying survival function.
Two approaches are commonly used to deal with failure time data with dependentcensoring (Hougaard,1986). One is the frailty approach that models the relationshipby using some latent variables (Zhang et al.,2005) and the other is the Copula modelapproach. In this paper, we will take the Copula model approach, which will bebriefy described in Section3.1along some notation. In Section3.2, we will presenta nonparametric estimation procedure for interval-censored data where the censoringvariables may be related to the failure time variable of interest. The procedure can be easily implemented and the key idea behind it is to divide the observed data into two sets of current status data, for which the estimation is relatively easy. Section3.3gives some numerical results and in particular, the method is illustrated by a set of interval-censored data arising from an AIDS study.
In this chapter, we will define the data structure and describe some notation and assumptions that will be used throughout the paper. In particular, the Copula model will be briefly discussed.
Consider a failure time study and let T denote the failure time of interest. Suppose that instead of observed exactly, the observation on T is characterized by two random variables U and V with U
Another advantage of the Copula model approach is its flexibility as there exist many different Copula functions.Among them,one class of the Copula functions that are commonly used is the Archimedean Copula functions with the three-dimensional function defined as C3(u1,u2,u3)=φ-1(φ(u1)+φ(u2)+φ(u3)),φ∈Φ,(7) whereΦdenotes the class of functionsφ:Ⅰ→[0,∞]with continuous first and second derivatives and satisfying φ(1)=0,φ'<0,φ">0,0
Specifically,let x0=0
引文
[1] ALIOUM, A. and COMMENGES, D.. A proportional hazards model for arbitrarilycensored and truncated data[J]. Biometrics,1996,52:512-524.
[2] BANERJEE, M. and WELLNER, J. A.. Likelihood ratio tests for monotone function-s[J]. Ann. Statist.,2001,29:1699-1731.
[3] BETENSKY, R. A. and FINKELSTEIN, D. M.. A non-parametric maximum likeli-hood estimator for bivariate interval censored data[J]. Statist. Med.,1999,18:3089-3100.
[4] BETENSKY,R.A.&FINKELSTEIN, D.M. Testing for dependence between failuretime and visit compliance with interval censored data [J]. Biometrics,2002,58:58-63.
[5] BHATTACHARYYA, AMIT. Modelling exponential survival data with dependent cen-soring[J]. The Indian Journal of Statistics,1997, Vol.59, Series A, No.2:242-267.
[6] BODHISATTVA SEN and MOULINATH BANERJEE. A pseudolikelihood methodfor analyzing interval censored data [J]. Biometrika,2007,94:71-86.
[7] Bo¨HNING, D., SCHLATTMANN, P. and Dietz, E.. Interval censored data: a noteon the nonparametric maximum likelihood estimator of the distribu-tion function[J].Biometrika,1996,83:462-466.
[8] BRAEKERS R. and VERAVERBEKE N.. A copula-graphic estimator for the condi-tional survival function under dependent censoring[J]. The Canadian Journal of Statis-tics,2005,33:429-447.
[9] BRAEKERS, R. and GADDAH, A.. A Koziol-Green estimator for the conditionaldistri-bution function under dependent censoring[J]. In Gomes M. I., Pestana D.,Silva P.(Eds). The56th Session of the International Statistical Institute-Proceedings.
[10] BRAEKERS, R. and GADDAH, A.. Flexible modeling in the Koziol-Green model bya copula function[J]. Communications in Statistics-Theory and Methods, In-press.
[11] BRAEKERS, R. and N. VERAVERKEKE (2008). The conditional Koziol-Green modelunder dependent censoring[J]. Statistics and Probability Letters,2008,78:927-937.
[12] BRUNEL, ELODIE and COMTE, FABIENNE. Cumulative distribution function es-timation under interval censoring case1[J]. Electronic Journal of Statistics,2009, Vol.3:1-24.
[13] CAI, J. and KIM, J.. Nonparametric quantile estimation with correlated failure timedata[J]. Lifetime Data Analysis,2003,9:357-371.
[14] CAI, J. and PRENTICE, R. L.. Estimating equations for hazard ratio parameters basedon correlated failure time data[J]. Biometrika,1995,82:151-164.
[15] Chen, Xiaohong, Fan, Yanqin. Estimation of copula-based semiparametric time seriesmodels Journal of Econometrics,2006,130(2):307-335.
[16] CHEN,DI, LU, JYE-CHYI and LIN, SHU CHUAN. Asymptotic distributions ofsemiparametric maximum likelihood estimators with estimating equations for group-censored data[J]. Aust. N. Z. J. Stat.,2005,47(2):173-192.
[17] Chen, Yi-Hau. Semiparametric marginal regression analysis for dependent competingrisks under an assumed copula[J]. Journal of the Royal Statistical Society: Series B(Statistical Methodology),2010,72(2):235-251.
[18] DEHGHAN, MOHAMMAD HOSSEIN and DUCHESNE, THIERRY. A generaliza-tion of Turnbull’s estimator for nonparametric estimation of the conditional survivalfunction with interval-censored data[J]. Lifetime Data Anal,2011,17:234-255.
[19] DEMPSTER, A.P., LAIRD, N.M., RUBIN, D.B.. Maximum likelihood from incompletedata via the EM algorithm[J]. J. Roy. Statist. Soc. Ser. B,,1977,39,1-22.
[20] DIAMOND, I.D., MCDONALD, J.W.. Analysis of current status data. In: Trussell, J.,Hankinson, R., Tilton, J.(Eds.), Demographic Applications of Event History Analysis.Oxford University Press, Oxford,1991, pp.231-252.
[21] DIAMOND, I.D., MCDONALD, J.W., SHAH, I.H.. Proportional hazards models forcurrent status data: application to the study of diferentials in age at weaning inPakistan[J]. Demography,1986,23:607-620.
[22] DIRIENZO, A. G.. Nonparametric comparison of two survival-time distributions inthe presence of dependent censoring[J]. Biometrics,2003,59:497-504.
[23] DUMBGEN, L., FREITAG-WOLF, S., JONGBLOEDF, G.. Estimating a unimodaldistribution from interval-censored data[J]. J. Amer. Statist. Assoc,2006,101:1094-1106.
[24] Dunson, D. B. and Dinse, G. E.. Bayesian models for multivariate current status datawith informative censoring[J]. Biometrics,2002,58:79-88.
[25] EBRAHIMI, NADER and MOLEFE, DANIEL. Survival function estimation whenlifetime and censoringtime are dependent[J]. Journal of Multivariate Analysis,2003,87:101-132.
[26] EMOTO, SHERRIE E. AND MATTHEWS, PETER C.. A weibull model for depen-dent censoring[J]. The Annals of Statistics,1990, Vol.18, No.4:1556-1577.
[27] FANG, HONGBIN, SUN, JIANGUO. Consistency of nonparametric maximum likeli-hood estimation of a distribution function based on doubly interval-censored failuretime data[J]. Statistics and Probability Letters,2001,55:311-318.
[28] FANG, H., SUN, J. LEE, M-L T. Nonparametric survival comparison for interval-censored continuous data[J]. Statistica Sinica,2002,12:1073-1083.
[29] FINE, J. P., JINAG, H. and CHAPPELL, R.. On semi-competing risks data[J].Biometrika,2001,88:907-919.
[30] FINKELSTEIN, D. M. A proportiional hazards model for interval censored failure timedata[J]. Biometrics,1986,42:845-854.
[31] FINKELSTEIN, D.M., WOLFE, R.A.. A semiparametric model for regression analysisof interval-censored failure time data[J]. Biometrics,1985,41:933-945.
[32] FINKELSTEIN, D.M., GEGGINS, W.B.&SCHOENFELD, D.A. Analysis of failuretime data with dependent interval censoring [J]. Biometrics,2002,58:298-304.
[33] FRYDMAN, H.. A note on nonparametric estimation of the distribution function frominterval-censored and truncated observations[J]. J R. Statist. Soc. B,1994,56:71-74.
[34] GENEST, C., MACKAY, J.. The joy of copulas: bivariate distribution with uniformmarginals[J]. Am Stat,1986,40:280-283.
[35] Genest, Christian, R′emillard, Bruno and Beaudoin, David. Goodness-of-ft tests forcopulas: a review and a power study[J]. Insurance: Mathematics and Economics,2009,44:199-213.
[36] GENTLEMAN, R. and VANDAL, A. C.. Computational algorithms for censored-dataproblems using intersection graphs[J]. J Comput. Graph. Statist.,2001,10:403-421.
[37] GENTLEMAN, R. and VANDAL, A. C.. Nonparametric estimation of the bivariateCDF for arbitrarily censored data[J]. Can. J Statist.,2002,30:557-571.
[38] GENTLEMAN, R.&GEYER, C. J. Maximum likelihood for interval censored data:consistency and computation [J]. Biometrika,1994,81:618-623.
[39] GO′MEZ, G.&LAGAKOS, S.W. Estimation of the infection time and latency distri-bution of AIDS with doubly censored data [J]. Biometrics,1994,50:204-212.
[40] GROENEBOOM, P.. Nonparametric maximum likelihood estimators for interval cen-soring and deconvolution. Technical Report378, Department of Statistics, StanfordUniversity,1991.
[41] GROENEBOOM, P., WELLNER, J.. Information bounds and nonparametric estima-tion[M]. DMV Seminar, Vol.19. Birkh¨auser Verlag, Berlin,1992.
[42] GRUTTOLA, D. V.&LAGAKOS,S. W. Analysis of doubly-censored survival data,with application to AIDS [J]. Biometrics,1989,45:1-11.
[43] GUSTAFSON, P., AESCHLIMAN, D. and LEVY, AR. A simple approach to fttingbayesian survival models. Lifetime Data Analysis,2003,9:5-19.
[44] HOEL, D. G.&WALBURG, H. E. Statistical analysis of survival experiments [M].Journal of National Cancer Institute,1972,49:361-372.
[45] HOUGAARD P. Fitting a multivariate failure time distribution. IEEE Trans Relia-bility,1989,38:444-448
[46] HOUGAARD, P. Analysis of multivariate survival data [M]. Springer-Verlag,New York,2000.
[47] HSU, CHIU-HSIEH and TAYLOR, JEREMY M. G.. Nonparametric comparison of twosurvival functions with dependent censoring via nonparametric multiple imputation[J].Statistics in Medicine,2009,28:462-475.
[48] HU, X. J.&LAWLESS, J. F. Estimation from truncated lifetime data with supplemen-tary information on covariates and censoring times [J]. Biometrika,1996,83:747-761.
[49] Huang, Xuelin and Wolfe, Robert A.. A Frailty Model for Informative Censoring[J]Biometrics,2002,58(3):510-520.
[50] HUDGENS, MICHAEL G.. On nonparametric maximum likelihood estimation withinterval censoring and left truncation [J]. J. R. Statist. Soc. B,2005,67, Part4, pp.573-587
[51] HUDGENS, M. G., SATTEN, G. A. and LONGINI, I. M.. Nonparametric maximumlikelihood estimation for competing risks survival data subject to interval censoringand truncation[J]. Biometrics,2001b,57:74-80.
[52] HU, X. J., LAWLESS, J. F.&SUZUKI, K. Nonparametric estimation of a lifetimedistribution when censoring times are missing [J]. Technometrics,1998,40:3-13.
[53] HUSTER, W.J., BROOKMEYER, R., SELF S.G.. Modeling paired survival data withcovariates[J]. Bio-metrics,1989,45:145-156.
[54] JAMMALAMADAKA, S.R., MANGALAM, V.. Non-parametric estimation formiddle-censored data[J]. J. Nonparametr. Statist,2003,15:253-265.
[55] JAMMALAMADAKA, S.R., IYER, S.K.. Approximate self consistency for middle-censored data[J]. J. Statist. Plann. Inference,2004,124:75-86.
[56] JEWELL, N. P. and VAN DER LAAN, M..Current status data: review, recent develop-ments and open problems[J]. Advances in survival analysis,2004,625-642, Handbookof Statist.,23, Elsevier, Amsterdam. MR2065792
[57] JIANG, HONGYU, FINE,JASON P., KOSOROK, MICHAEL R. and CHAPPELL,RICK. Pseudo self-consistent estimation of a copula model with informative censor-ing[J]. The Scandinavian Journal of Statistics,2005,32:1-20.
[58] JOE, HARRY. Multivariate models and dependence concepts [M]. First edition, Chap-man&Hall, Great Britain by St Edmundsbury press, Bury St Edmunds,Sufolk,1997.
[59] JONGBLOED, G.. The iterative convex minorant algorithm for nonparametric esti-mation[J]. J. Comput. Graphical Statist,1998,7:301-321.
[60] JULIAà, OLGA and GóMEZ, GUADALUPE. Simultaneous marginal survival esti-mators when doubly censored data is present[J]. Lifetime Data Anal,2011,17:347-372.
[61] KALBFLEISCH, J. D.&PRENTICE, R. L. The statistical analysis of failure timedata [M]. Second edition, John Wiley: New York,2002.
[62] KAPLAN, E.L., MEIER, P.. Nonparametric estimation from incomplete observation-s[J]. J. Amer. Statist. Assoc.1958,53,457-481. MR0093867
[63] KEIDING, N.. Age-specifc incidence and prevalence: a statistical perspective (withdiscussion)[J]. J. Roy. Statist. Soc. Ser. A,1991,154:371-412.
[64] KLEIN, J. P.&MOESCHBERGER, M. L. Survival Analysis [M]. Springer-Verlag:New York,2003.
[65] KOZIOL J.A. and GREEN, S.B.. A Crame′r-von Mises statistic for randomly censoreddata[J]. Biometrika,1976,63:465-474.
[66] LAGAKOS, S. W.&LOUIS, T. A. Use of tumor lethality to interpret tumorigenicityexperiments lacking cause-of-death data [J]. Applied Statistics,1988,37:169-179.
[67] LI, L.&PU. Z. Regression models with arbitrarily interval-censored observations [J].Communications in Statistics. Theory and Methods,1999,28:1547-1563.
[68] LI, L.&PU. Z. Rank estimation of log-linear regression with interval-censored data[J]. Lifetime Data Analysis,2003,9:57-70.
[69] Li, Yi,Tiwari, Ram C. and Guha, Subharup. Mixture cure survival models withdependent censoring[J]. J. R. Statist. Soc. B,2007,69, Part3:285-306.
[70] LO, SIMON M. S and WILKE, RALF A. A copula model for dependent competingrisks[J]. Journal of the Royal Statistical Society. Series C (Applied Statistics),2010,59(2):359-376.
[71] LIN, D.Y., ROBINS, J.M. ADN WEI, L.J.. Comparing two failure time distributionsin the presense of dependent censoring[J]. Biometrika,1996,83,381-393.
[72] LONG, XIANGDONG.Multivariate GARCH models using Copula, nonparametric andsemiparametric methods[M].The University of California(Riverside) PH.D. disserta-tion,2005.
[73] MA, S. and KOSOROK, M.R.. Adaptive penalized M-estimation with current statusdata[J]. Ann. Inst. Statist. Math.2006,58,511-526. MR2327890
[74] MCKEOWN, K.&JEWELL, N. P. Misclassifcation of current status data [J]. LifetimeData Analysis,2010,16:215-230.
[75] MODARRES, REZA. A test of independence based on the likelihood of cut-points[J].Communications in Statistics―Simulation and Computation,2007,36:817-825.
[76] NELSEN, R. B. An introduction to copulas [M]. Second Edition. Springer-Verlag,NewYork,2006.
[77] OAKES, D. Bivariate survival models induced by frailties. J Am Statist Assoc,1989,84:487-493.
[78] PARK,Yongseok, KALBFLEISCH, John D. and TAYLOR,Jeremy M. G.. Con-strained nonparametric maximum likelihood estimation of stochastically ordered sur-vivor functions[J]. The Canadian Journal of Statistics,2012, Vol.40, No.1:22-39
[79] PETO, R. Experimental survival curves for interval-censored data[J]. Appl. S-tatist.,1973,22:86-91.
[80] RIVEST L. and WELLS, M.T.. A martingale approach to the copula-graphic estima-tor for the survival function under dependent censoring[J]. J. Multivariate Analysis2001,79:138-155.
[81] ROMEO,JOSe′S., TANAKA, NELSON I., PEDROSO-DE-LIMA, ANTONIO C..Bivariate survival modeling: a bayesian approach based on copulas[J]. Lifetime DataAnal,2006,12:205-222.
[82] SA¨ D, M′eRIEM and GHAZZALI, NADIA and RIVEST, LOUIS PAUL. Score tests fordependent censoring with survival data[J]. Lecture Notes Monograph Series, Vol.42,Mathematical Statistics and Applications: Festschrift for Constance van Eeden,2003,pp435-461.
[83] SCHICK, A. and YU, Q.. Consistency of the GMLE with mixed case interval-censoreddata[J]. Scand. J Statist.,2000,27:45-55.
[84] SCHWEIZER B, SKLAR A. Probabilistic metric spaces[M]. North-Holland/Elsevier,New York, NY,1983.
[85] SHIH, JH, LOUIS, TA. Inferences on the association parameter in copula models forbivariate survival data. Biometrics,1995,51:1384-1399.
[86] SHEN, PAO-SHENG. Nonparametric estimation of the bivariate survival function forone modifed form of doubly censored data[J]. Comput Stat.,2010,25:203-213.
[87] SHEN, PAO-SHENG. An inverse-probability-weighted approach to the estimation ofdistribution function with middle-censored data[J]. Journal of Statistical Planning andInference,2010,140:1844-1851.
[88] SHEN, XIAOJING, ZHUA, YUNMIN, SONG, LIXIN. Linear B-spline copulas withapplications to nonparametric estimation of copulas[J]. Computational Statistics andData Analysis,2008,52:3806-3819.
[89] SKLAR, A. Fonctions de re′partition a`n dimensions et leurs marges. Publ Inst StatistUniv Paris,1959,8:229-231.
[90] SLUD, ERIC V. and RUBINSTEIN, LAWRENCE V.. Dependent competing risks andsummary survival Curves[J]. Biometrika,1983,70(3):643-649.
[91] SONG, S.. Estimation with univariate mixed case interval censored data[J]. Statist.Sin.,2004,14:269-282.
[92] SUGIMOTO, TOMOYUKI. A Wald-type variance estimation for the nonparametricdistribution estimators for doubly censored data[J]. Ann Inst Stat Math,2011,63:645-670.
[93] SUN, J. Interval censoring [M]. Encyclopedia of Biostatistics, John Wiley, Second Edi-tion,2005,2603-2609.
[94] SUN, J. The statistical analysis of interval-censored failure time data [M]. New York:Springer,2006.
[95] SUN, J.&SUN, L. Semiparametric linear transformation models for current statusdata [J]. The Canadian Journal of Statistics,2005,33:85-96.
[96] TURNBULL, B.W.. Nonparametric estimation of a survivorship function with doublycensored data[J]. J. Amer. Statist. Assoc.,1974,69:169-173.
[97] TURNBULL, B. W. The empirical distribution function with arbitrarily grouped, cen-sored and truncated data [J]. Journal of the Royal Statistical Society: Series B,1976,38:290-295.
[98] TURNBULL, B. W. and WEISS, L.. A likelihood ratio statistic for testing goodnessof ft with randomly censored data[J]. Biometrics,1978,34,367-375.
[99] VAN DE GEER, S.. Hellinger-consistency of certain nonparametric like-lihood esti-mators[J]. Ann. Statist.1993,21,14-44. MR1212164
[100] VAV DER VAART, A. and VAN DER LAAN, M. J.. Estimating a survival distributionwith current status data and high-dimensional covariates[J]. Int. J. Biostat.2006,2,Art9,42pp. MR2306498
[101] VAV DER VAART, A. and WELLNER, J. A.. Preservation theorems for Glivenko-Cantelli and uniform Glivenko-Cantelli classe[J]. In High Dimensional Probability II,2000, pp.115-133. Boston: Birkhiuser.
[102] VERAVERBEKE, N. and CADARSO SU′aREZ, C.. Estimation of the conditionaldistribution in a conditional Koziol-Green model[J]. Test,2000,9:97-122.
[103] WANG, WEIJING. Estimating the association parameter for copula models underdependent censoring[J]. J. R. Statist. Soc. B,2003,65, Part1:257-273.
[104] WANG, YONG. Dimension-reduced nonparametric maximum likelihood computationfor interval-censored data[J]. Computational Statistics and Data Analysis,2008,52:2388-2402.
[105] WELLNER, J.A.&ZHAN, YIHUI. A hybrid algorithm for computation of the non-parametric maximum Likelihood estimator from censored data [J]. Journal of the Amer-ican Statistical Association,1997,92:945-959.
[106] WONG, G. Y. C. and YU, Q.. Generalized MLE of a joint distribution function withmultivariate interval-censored data. J Multiv. Anal.,1999,69:155-166.
[107] YANG, SONG. Functional estimation under interval censoring case1[J]. Journal ofStatistical Planning and Inference2000,89:135-144.
[108] YU, Q., LI, L. and WONG, G. Y. C.. Asymptotic variance of the GMLE of a survivalfunction with interval-censored data[J]. Sankhya,1998a,60:184-187.
[109] YU, Q., SCHICK, A., LI, L. and WONG, G. Y. C.. Asymptotic properties of theGMLE with case2interval-censored data. Statist[J]. Probab. Lett.,1998b,37:223-228.
[110] ZHANG, YING and JAMSHIDIAN, MORTAZA. On algorithms for the nonparametricmaximum likelihood estimator of the failure function with censored data[J]. J. Comput.Graphical Statist,2004,13:123-140.
[111] ZHANG, ZHIGANG&SUN, JIANGUO. Interval censoring [J]. Statistical Methods inMedical Research,2010,19:53-70.
[112] ZHANG, Z., SUN, J.&SUN L. Statistical analysis of current status data with infor-mative observation times [J]. Statistics in Medicine,2005,24:1399-1407.
[113] ZHANG, ZHIGANG,SUN, LIUQUAN, Sun,JIANGUO and FINKELSTEIN,D.M.Regression analysis of failure time data with informative interval censoring [J]. Statisticsin Medicine,2007,26:2533-2546.
[114] ZHENG, M.&KLEIN, J. P. Estimates of marginal survival for dependent competingrisk based on an assumed copula.[J]. Biometrika,1995,82:127-38.
[115] ZHENG, M. On the use of copulas in dependent competing risks theory [M] The OhioState University PH.D. dissertation,1992.
[116] ZHU, L., TONG, X. and SUN, J. A transformation approach for the analysis of interval-censored failure time data [J]. Lifetime Data Analysis,2008,14:167-178.
[2] BANERJEE, M. and WELLNER, J. A.. Likelihood ratio tests for monotone function-s[J]. Ann. Statist.,2001,29:1699-1731.
[3] BETENSKY, R. A. and FINKELSTEIN, D. M.. A non-parametric maximum likeli-hood estimator for bivariate interval censored data[J]. Statist. Med.,1999,18:3089-3100.
[4] BETENSKY,R.A.&FINKELSTEIN, D.M. Testing for dependence between failuretime and visit compliance with interval censored data [J]. Biometrics,2002,58:58-63.
[5] BHATTACHARYYA, AMIT. Modelling exponential survival data with dependent cen-soring[J]. The Indian Journal of Statistics,1997, Vol.59, Series A, No.2:242-267.
[6] BODHISATTVA SEN and MOULINATH BANERJEE. A pseudolikelihood methodfor analyzing interval censored data [J]. Biometrika,2007,94:71-86.
[7] Bo¨HNING, D., SCHLATTMANN, P. and Dietz, E.. Interval censored data: a noteon the nonparametric maximum likelihood estimator of the distribu-tion function[J].Biometrika,1996,83:462-466.
[8] BRAEKERS R. and VERAVERBEKE N.. A copula-graphic estimator for the condi-tional survival function under dependent censoring[J]. The Canadian Journal of Statis-tics,2005,33:429-447.
[9] BRAEKERS, R. and GADDAH, A.. A Koziol-Green estimator for the conditionaldistri-bution function under dependent censoring[J]. In Gomes M. I., Pestana D.,Silva P.(Eds). The56th Session of the International Statistical Institute-Proceedings.
[10] BRAEKERS, R. and GADDAH, A.. Flexible modeling in the Koziol-Green model bya copula function[J]. Communications in Statistics-Theory and Methods, In-press.
[11] BRAEKERS, R. and N. VERAVERKEKE (2008). The conditional Koziol-Green modelunder dependent censoring[J]. Statistics and Probability Letters,2008,78:927-937.
[12] BRUNEL, ELODIE and COMTE, FABIENNE. Cumulative distribution function es-timation under interval censoring case1[J]. Electronic Journal of Statistics,2009, Vol.3:1-24.
[13] CAI, J. and KIM, J.. Nonparametric quantile estimation with correlated failure timedata[J]. Lifetime Data Analysis,2003,9:357-371.
[14] CAI, J. and PRENTICE, R. L.. Estimating equations for hazard ratio parameters basedon correlated failure time data[J]. Biometrika,1995,82:151-164.
[15] Chen, Xiaohong, Fan, Yanqin. Estimation of copula-based semiparametric time seriesmodels Journal of Econometrics,2006,130(2):307-335.
[16] CHEN,DI, LU, JYE-CHYI and LIN, SHU CHUAN. Asymptotic distributions ofsemiparametric maximum likelihood estimators with estimating equations for group-censored data[J]. Aust. N. Z. J. Stat.,2005,47(2):173-192.
[17] Chen, Yi-Hau. Semiparametric marginal regression analysis for dependent competingrisks under an assumed copula[J]. Journal of the Royal Statistical Society: Series B(Statistical Methodology),2010,72(2):235-251.
[18] DEHGHAN, MOHAMMAD HOSSEIN and DUCHESNE, THIERRY. A generaliza-tion of Turnbull’s estimator for nonparametric estimation of the conditional survivalfunction with interval-censored data[J]. Lifetime Data Anal,2011,17:234-255.
[19] DEMPSTER, A.P., LAIRD, N.M., RUBIN, D.B.. Maximum likelihood from incompletedata via the EM algorithm[J]. J. Roy. Statist. Soc. Ser. B,,1977,39,1-22.
[20] DIAMOND, I.D., MCDONALD, J.W.. Analysis of current status data. In: Trussell, J.,Hankinson, R., Tilton, J.(Eds.), Demographic Applications of Event History Analysis.Oxford University Press, Oxford,1991, pp.231-252.
[21] DIAMOND, I.D., MCDONALD, J.W., SHAH, I.H.. Proportional hazards models forcurrent status data: application to the study of diferentials in age at weaning inPakistan[J]. Demography,1986,23:607-620.
[22] DIRIENZO, A. G.. Nonparametric comparison of two survival-time distributions inthe presence of dependent censoring[J]. Biometrics,2003,59:497-504.
[23] DUMBGEN, L., FREITAG-WOLF, S., JONGBLOEDF, G.. Estimating a unimodaldistribution from interval-censored data[J]. J. Amer. Statist. Assoc,2006,101:1094-1106.
[24] Dunson, D. B. and Dinse, G. E.. Bayesian models for multivariate current status datawith informative censoring[J]. Biometrics,2002,58:79-88.
[25] EBRAHIMI, NADER and MOLEFE, DANIEL. Survival function estimation whenlifetime and censoringtime are dependent[J]. Journal of Multivariate Analysis,2003,87:101-132.
[26] EMOTO, SHERRIE E. AND MATTHEWS, PETER C.. A weibull model for depen-dent censoring[J]. The Annals of Statistics,1990, Vol.18, No.4:1556-1577.
[27] FANG, HONGBIN, SUN, JIANGUO. Consistency of nonparametric maximum likeli-hood estimation of a distribution function based on doubly interval-censored failuretime data[J]. Statistics and Probability Letters,2001,55:311-318.
[28] FANG, H., SUN, J. LEE, M-L T. Nonparametric survival comparison for interval-censored continuous data[J]. Statistica Sinica,2002,12:1073-1083.
[29] FINE, J. P., JINAG, H. and CHAPPELL, R.. On semi-competing risks data[J].Biometrika,2001,88:907-919.
[30] FINKELSTEIN, D. M. A proportiional hazards model for interval censored failure timedata[J]. Biometrics,1986,42:845-854.
[31] FINKELSTEIN, D.M., WOLFE, R.A.. A semiparametric model for regression analysisof interval-censored failure time data[J]. Biometrics,1985,41:933-945.
[32] FINKELSTEIN, D.M., GEGGINS, W.B.&SCHOENFELD, D.A. Analysis of failuretime data with dependent interval censoring [J]. Biometrics,2002,58:298-304.
[33] FRYDMAN, H.. A note on nonparametric estimation of the distribution function frominterval-censored and truncated observations[J]. J R. Statist. Soc. B,1994,56:71-74.
[34] GENEST, C., MACKAY, J.. The joy of copulas: bivariate distribution with uniformmarginals[J]. Am Stat,1986,40:280-283.
[35] Genest, Christian, R′emillard, Bruno and Beaudoin, David. Goodness-of-ft tests forcopulas: a review and a power study[J]. Insurance: Mathematics and Economics,2009,44:199-213.
[36] GENTLEMAN, R. and VANDAL, A. C.. Computational algorithms for censored-dataproblems using intersection graphs[J]. J Comput. Graph. Statist.,2001,10:403-421.
[37] GENTLEMAN, R. and VANDAL, A. C.. Nonparametric estimation of the bivariateCDF for arbitrarily censored data[J]. Can. J Statist.,2002,30:557-571.
[38] GENTLEMAN, R.&GEYER, C. J. Maximum likelihood for interval censored data:consistency and computation [J]. Biometrika,1994,81:618-623.
[39] GO′MEZ, G.&LAGAKOS, S.W. Estimation of the infection time and latency distri-bution of AIDS with doubly censored data [J]. Biometrics,1994,50:204-212.
[40] GROENEBOOM, P.. Nonparametric maximum likelihood estimators for interval cen-soring and deconvolution. Technical Report378, Department of Statistics, StanfordUniversity,1991.
[41] GROENEBOOM, P., WELLNER, J.. Information bounds and nonparametric estima-tion[M]. DMV Seminar, Vol.19. Birkh¨auser Verlag, Berlin,1992.
[42] GRUTTOLA, D. V.&LAGAKOS,S. W. Analysis of doubly-censored survival data,with application to AIDS [J]. Biometrics,1989,45:1-11.
[43] GUSTAFSON, P., AESCHLIMAN, D. and LEVY, AR. A simple approach to fttingbayesian survival models. Lifetime Data Analysis,2003,9:5-19.
[44] HOEL, D. G.&WALBURG, H. E. Statistical analysis of survival experiments [M].Journal of National Cancer Institute,1972,49:361-372.
[45] HOUGAARD P. Fitting a multivariate failure time distribution. IEEE Trans Relia-bility,1989,38:444-448
[46] HOUGAARD, P. Analysis of multivariate survival data [M]. Springer-Verlag,New York,2000.
[47] HSU, CHIU-HSIEH and TAYLOR, JEREMY M. G.. Nonparametric comparison of twosurvival functions with dependent censoring via nonparametric multiple imputation[J].Statistics in Medicine,2009,28:462-475.
[48] HU, X. J.&LAWLESS, J. F. Estimation from truncated lifetime data with supplemen-tary information on covariates and censoring times [J]. Biometrika,1996,83:747-761.
[49] Huang, Xuelin and Wolfe, Robert A.. A Frailty Model for Informative Censoring[J]Biometrics,2002,58(3):510-520.
[50] HUDGENS, MICHAEL G.. On nonparametric maximum likelihood estimation withinterval censoring and left truncation [J]. J. R. Statist. Soc. B,2005,67, Part4, pp.573-587
[51] HUDGENS, M. G., SATTEN, G. A. and LONGINI, I. M.. Nonparametric maximumlikelihood estimation for competing risks survival data subject to interval censoringand truncation[J]. Biometrics,2001b,57:74-80.
[52] HU, X. J., LAWLESS, J. F.&SUZUKI, K. Nonparametric estimation of a lifetimedistribution when censoring times are missing [J]. Technometrics,1998,40:3-13.
[53] HUSTER, W.J., BROOKMEYER, R., SELF S.G.. Modeling paired survival data withcovariates[J]. Bio-metrics,1989,45:145-156.
[54] JAMMALAMADAKA, S.R., MANGALAM, V.. Non-parametric estimation formiddle-censored data[J]. J. Nonparametr. Statist,2003,15:253-265.
[55] JAMMALAMADAKA, S.R., IYER, S.K.. Approximate self consistency for middle-censored data[J]. J. Statist. Plann. Inference,2004,124:75-86.
[56] JEWELL, N. P. and VAN DER LAAN, M..Current status data: review, recent develop-ments and open problems[J]. Advances in survival analysis,2004,625-642, Handbookof Statist.,23, Elsevier, Amsterdam. MR2065792
[57] JIANG, HONGYU, FINE,JASON P., KOSOROK, MICHAEL R. and CHAPPELL,RICK. Pseudo self-consistent estimation of a copula model with informative censor-ing[J]. The Scandinavian Journal of Statistics,2005,32:1-20.
[58] JOE, HARRY. Multivariate models and dependence concepts [M]. First edition, Chap-man&Hall, Great Britain by St Edmundsbury press, Bury St Edmunds,Sufolk,1997.
[59] JONGBLOED, G.. The iterative convex minorant algorithm for nonparametric esti-mation[J]. J. Comput. Graphical Statist,1998,7:301-321.
[60] JULIAà, OLGA and GóMEZ, GUADALUPE. Simultaneous marginal survival esti-mators when doubly censored data is present[J]. Lifetime Data Anal,2011,17:347-372.
[61] KALBFLEISCH, J. D.&PRENTICE, R. L. The statistical analysis of failure timedata [M]. Second edition, John Wiley: New York,2002.
[62] KAPLAN, E.L., MEIER, P.. Nonparametric estimation from incomplete observation-s[J]. J. Amer. Statist. Assoc.1958,53,457-481. MR0093867
[63] KEIDING, N.. Age-specifc incidence and prevalence: a statistical perspective (withdiscussion)[J]. J. Roy. Statist. Soc. Ser. A,1991,154:371-412.
[64] KLEIN, J. P.&MOESCHBERGER, M. L. Survival Analysis [M]. Springer-Verlag:New York,2003.
[65] KOZIOL J.A. and GREEN, S.B.. A Crame′r-von Mises statistic for randomly censoreddata[J]. Biometrika,1976,63:465-474.
[66] LAGAKOS, S. W.&LOUIS, T. A. Use of tumor lethality to interpret tumorigenicityexperiments lacking cause-of-death data [J]. Applied Statistics,1988,37:169-179.
[67] LI, L.&PU. Z. Regression models with arbitrarily interval-censored observations [J].Communications in Statistics. Theory and Methods,1999,28:1547-1563.
[68] LI, L.&PU. Z. Rank estimation of log-linear regression with interval-censored data[J]. Lifetime Data Analysis,2003,9:57-70.
[69] Li, Yi,Tiwari, Ram C. and Guha, Subharup. Mixture cure survival models withdependent censoring[J]. J. R. Statist. Soc. B,2007,69, Part3:285-306.
[70] LO, SIMON M. S and WILKE, RALF A. A copula model for dependent competingrisks[J]. Journal of the Royal Statistical Society. Series C (Applied Statistics),2010,59(2):359-376.
[71] LIN, D.Y., ROBINS, J.M. ADN WEI, L.J.. Comparing two failure time distributionsin the presense of dependent censoring[J]. Biometrika,1996,83,381-393.
[72] LONG, XIANGDONG.Multivariate GARCH models using Copula, nonparametric andsemiparametric methods[M].The University of California(Riverside) PH.D. disserta-tion,2005.
[73] MA, S. and KOSOROK, M.R.. Adaptive penalized M-estimation with current statusdata[J]. Ann. Inst. Statist. Math.2006,58,511-526. MR2327890
[74] MCKEOWN, K.&JEWELL, N. P. Misclassifcation of current status data [J]. LifetimeData Analysis,2010,16:215-230.
[75] MODARRES, REZA. A test of independence based on the likelihood of cut-points[J].Communications in Statistics―Simulation and Computation,2007,36:817-825.
[76] NELSEN, R. B. An introduction to copulas [M]. Second Edition. Springer-Verlag,NewYork,2006.
[77] OAKES, D. Bivariate survival models induced by frailties. J Am Statist Assoc,1989,84:487-493.
[78] PARK,Yongseok, KALBFLEISCH, John D. and TAYLOR,Jeremy M. G.. Con-strained nonparametric maximum likelihood estimation of stochastically ordered sur-vivor functions[J]. The Canadian Journal of Statistics,2012, Vol.40, No.1:22-39
[79] PETO, R. Experimental survival curves for interval-censored data[J]. Appl. S-tatist.,1973,22:86-91.
[80] RIVEST L. and WELLS, M.T.. A martingale approach to the copula-graphic estima-tor for the survival function under dependent censoring[J]. J. Multivariate Analysis2001,79:138-155.
[81] ROMEO,JOSe′S., TANAKA, NELSON I., PEDROSO-DE-LIMA, ANTONIO C..Bivariate survival modeling: a bayesian approach based on copulas[J]. Lifetime DataAnal,2006,12:205-222.
[82] SA¨ D, M′eRIEM and GHAZZALI, NADIA and RIVEST, LOUIS PAUL. Score tests fordependent censoring with survival data[J]. Lecture Notes Monograph Series, Vol.42,Mathematical Statistics and Applications: Festschrift for Constance van Eeden,2003,pp435-461.
[83] SCHICK, A. and YU, Q.. Consistency of the GMLE with mixed case interval-censoreddata[J]. Scand. J Statist.,2000,27:45-55.
[84] SCHWEIZER B, SKLAR A. Probabilistic metric spaces[M]. North-Holland/Elsevier,New York, NY,1983.
[85] SHIH, JH, LOUIS, TA. Inferences on the association parameter in copula models forbivariate survival data. Biometrics,1995,51:1384-1399.
[86] SHEN, PAO-SHENG. Nonparametric estimation of the bivariate survival function forone modifed form of doubly censored data[J]. Comput Stat.,2010,25:203-213.
[87] SHEN, PAO-SHENG. An inverse-probability-weighted approach to the estimation ofdistribution function with middle-censored data[J]. Journal of Statistical Planning andInference,2010,140:1844-1851.
[88] SHEN, XIAOJING, ZHUA, YUNMIN, SONG, LIXIN. Linear B-spline copulas withapplications to nonparametric estimation of copulas[J]. Computational Statistics andData Analysis,2008,52:3806-3819.
[89] SKLAR, A. Fonctions de re′partition a`n dimensions et leurs marges. Publ Inst StatistUniv Paris,1959,8:229-231.
[90] SLUD, ERIC V. and RUBINSTEIN, LAWRENCE V.. Dependent competing risks andsummary survival Curves[J]. Biometrika,1983,70(3):643-649.
[91] SONG, S.. Estimation with univariate mixed case interval censored data[J]. Statist.Sin.,2004,14:269-282.
[92] SUGIMOTO, TOMOYUKI. A Wald-type variance estimation for the nonparametricdistribution estimators for doubly censored data[J]. Ann Inst Stat Math,2011,63:645-670.
[93] SUN, J. Interval censoring [M]. Encyclopedia of Biostatistics, John Wiley, Second Edi-tion,2005,2603-2609.
[94] SUN, J. The statistical analysis of interval-censored failure time data [M]. New York:Springer,2006.
[95] SUN, J.&SUN, L. Semiparametric linear transformation models for current statusdata [J]. The Canadian Journal of Statistics,2005,33:85-96.
[96] TURNBULL, B.W.. Nonparametric estimation of a survivorship function with doublycensored data[J]. J. Amer. Statist. Assoc.,1974,69:169-173.
[97] TURNBULL, B. W. The empirical distribution function with arbitrarily grouped, cen-sored and truncated data [J]. Journal of the Royal Statistical Society: Series B,1976,38:290-295.
[98] TURNBULL, B. W. and WEISS, L.. A likelihood ratio statistic for testing goodnessof ft with randomly censored data[J]. Biometrics,1978,34,367-375.
[99] VAN DE GEER, S.. Hellinger-consistency of certain nonparametric like-lihood esti-mators[J]. Ann. Statist.1993,21,14-44. MR1212164
[100] VAV DER VAART, A. and VAN DER LAAN, M. J.. Estimating a survival distributionwith current status data and high-dimensional covariates[J]. Int. J. Biostat.2006,2,Art9,42pp. MR2306498
[101] VAV DER VAART, A. and WELLNER, J. A.. Preservation theorems for Glivenko-Cantelli and uniform Glivenko-Cantelli classe[J]. In High Dimensional Probability II,2000, pp.115-133. Boston: Birkhiuser.
[102] VERAVERBEKE, N. and CADARSO SU′aREZ, C.. Estimation of the conditionaldistribution in a conditional Koziol-Green model[J]. Test,2000,9:97-122.
[103] WANG, WEIJING. Estimating the association parameter for copula models underdependent censoring[J]. J. R. Statist. Soc. B,2003,65, Part1:257-273.
[104] WANG, YONG. Dimension-reduced nonparametric maximum likelihood computationfor interval-censored data[J]. Computational Statistics and Data Analysis,2008,52:2388-2402.
[105] WELLNER, J.A.&ZHAN, YIHUI. A hybrid algorithm for computation of the non-parametric maximum Likelihood estimator from censored data [J]. Journal of the Amer-ican Statistical Association,1997,92:945-959.
[106] WONG, G. Y. C. and YU, Q.. Generalized MLE of a joint distribution function withmultivariate interval-censored data. J Multiv. Anal.,1999,69:155-166.
[107] YANG, SONG. Functional estimation under interval censoring case1[J]. Journal ofStatistical Planning and Inference2000,89:135-144.
[108] YU, Q., LI, L. and WONG, G. Y. C.. Asymptotic variance of the GMLE of a survivalfunction with interval-censored data[J]. Sankhya,1998a,60:184-187.
[109] YU, Q., SCHICK, A., LI, L. and WONG, G. Y. C.. Asymptotic properties of theGMLE with case2interval-censored data. Statist[J]. Probab. Lett.,1998b,37:223-228.
[110] ZHANG, YING and JAMSHIDIAN, MORTAZA. On algorithms for the nonparametricmaximum likelihood estimator of the failure function with censored data[J]. J. Comput.Graphical Statist,2004,13:123-140.
[111] ZHANG, ZHIGANG&SUN, JIANGUO. Interval censoring [J]. Statistical Methods inMedical Research,2010,19:53-70.
[112] ZHANG, Z., SUN, J.&SUN L. Statistical analysis of current status data with infor-mative observation times [J]. Statistics in Medicine,2005,24:1399-1407.
[113] ZHANG, ZHIGANG,SUN, LIUQUAN, Sun,JIANGUO and FINKELSTEIN,D.M.Regression analysis of failure time data with informative interval censoring [J]. Statisticsin Medicine,2007,26:2533-2546.
[114] ZHENG, M.&KLEIN, J. P. Estimates of marginal survival for dependent competingrisk based on an assumed copula.[J]. Biometrika,1995,82:127-38.
[115] ZHENG, M. On the use of copulas in dependent competing risks theory [M] The OhioState University PH.D. dissertation,1992.
[116] ZHU, L., TONG, X. and SUN, J. A transformation approach for the analysis of interval-censored failure time data [J]. Lifetime Data Analysis,2008,14:167-178.