A common objective in longitudinal studies is to characterize the relationship between a longitudinal response process and a time-to-event data. Ordinal nature of the response and possible missing information on covariates add complications to the joint model. In such circumstances, some influential observations often present in the data may upset the analysis. In this paper, a joint model based on ordinal partial mixed model and an accelerated failure time model is used, to account for the repeated ordered response and time-to-event data, respectively. Here, we propose an influence function-based robust estimation method. Monte Carlo expectation maximization method-based algorithm is used for parameter estimation. A detailed simulation study has been done to evaluate the performance of the proposed method. As an application, a data on muscular dystrophy among children is used. Robust estimates are then compared with classical maximum likelihood estimates.
WangYTaylorJMG. Jointly modelling longitudinal and event time data with application to aids studies. J Am Stat Assoc2001; 96: 895–905.
2.
TaylorJMGCumberlandWGSyJP. A stochastic model for analysis of longitudinal aids data. J Am Stat Assoc1994; 89: 727–736.
3.
LavalleyMPDe GruttolaV. Models for empirical Bayes estimators of longitudinal cd4 counts (disc: P2337-2340). Stat Med1996; 15: 2289–2305.
4.
HendersonRDigglePJDobsonA. Joint modeling of longitudinal measurements and event time data. Biostatistics2000; 1: 465–480.
5.
XuJZegerSL. Joint analysis of longitudinal data comprising repeated measures and times to events. J R Stat Soc Ser C: Appl Stat2001; 50(3): 375–387.
6.
EngleRFGrangerWJRiceJet al.Semiparametric estimates of the relation between weather and electricity sales. J Am Stat Assoc1986; 80: 310–319.
7.
ClevelandWS. Robust locally-weighted regression and smoothing scatterplots. J Am Stat Assoc1979; 74: 829–836.
8.
HastieTJTibshiraniRJ. Generalized additive models (with discussion). Stat Sci1986; 1(2): 297–318.
9.
SilvermanBW. Spline smoothing: The equivalent variable kernel method. Ann Stat1984; 12: 898–916.
10.
SpeckmanPE. Regression analysis for partially linear models. J R Stat Soc Ser B1988; 50: 413–436.
11.
MullerM. Estimation and testing in generalized partial linear models – a comparative study. Stat Comput2001; 11: 299–309.
12.
BaiYZhuZFungW. Partial linear models for longitudinal data based on quadratic inference functions. Scand J Stat2008; 35: 104118.
13.
ZegerSDiggleP. Semiparametric models for longitudinal data with application to cd4 cell numbers in HIV seroconverters. Biometrics1994; 50: 689–699.
14.
HeXFungWKZhuZ. Robust estimation in generalized partial linear models for clustered data. J Am Stat Assoc2005; 100: 1176–1184.
15.
HsiehFTsengY-KWangJ-L. Joint modeling of survival and longitudinal data: Likelihood approach revisited. Biometrics2006; 62: 1037–1043.
16.
TsiatisAADavidianM. Joint modeling of longitudinal and time-to-event data: An overview. Stat Sin2004; 14: 809–834.
17.
KeidingNAndersenPKKleinJP. The role of frailty models and accelerated failure time models in describing heterogeneity due to omitted covariates. Stat Med1997; 16: 215–224.
18.
LambertPCollettDKimberAet al.Parametric accelerated failure time models with random effects and an application to kidney transplant survival. Stat Med2004; 23: 3177–3192.
19.
RubinDB. Inference and missing data. Biometrika1976; 63: 581–592.
20.
WuL. A joint model for nonlinear mixed-effects models with censoring and covariates measured with error, with application to aids studies. J Am Stat Assoc2002; 97: 955–964.
21.
WuLHuJWuH. Joint inference for nonlinear mixed-effects models and time-to-event at the presence of missing data. Biostatistics2007; 9: 308–320.
22.
LiangHQinY. Empirical likelihood-based inferences for partially linear models with missing covariates. Aust NZ J Stat2008; 50: 347–359.
23.
LiangHWangSRobinsJMet al.Estimation in partially linear models with missing covariates. J Am Stat Assoc2004; 99: 357–367.
24.
RobinsJMRotnitzkyAZhaoLP. Estimation of regression coefficients when some regressors are not always observed. J Am Stat Assoc1994; 89: 846–866.
25.
CantoniERonchettiE. Robust inference for generalized linear models. J Am Stat Assoc2001; 96: 1022–1030.
26.
CoptSVictoria-FeserMP. High breakdown inference in the mixed linear model. J Am Stat Assoc2006; 101: 292–300.
27.
PreisserJSQaquishBF. Robust regression to cluster data with application to binary responses. Biometrics1999; 55: 574–579.
28.
RichardsonAM. Bounded influence estimation in the mixed linear model. J Am Stat Assoc1997; 92: 154–161.
29.
MoustakiIVictoria-FeserMP. Bounded-influence robust estimation in generalized linear latent variable models. J Am Stat Assoc2006; 101: 644–653.
30.
QinGYZhuZY. Robust estimation in generalized semiparametric mixed model for longitudinal data. J Multivariate Anal2007; 98: 1658–1683.
31.
WuLQiuJ. Approximate bounded influence estimation for longitudinal data with outliers and measurement errors. J Stat Plan Inference2011; 141(7): 2321–2330.
32.
HampelFRRonchettiEMRousseeuwPJet al.Robust statistics. The approach based on influence functions, New York: Wiley, 1986.
33.
HampelFR. The influence curve and its role in robust estimation. J Am Stat Assoc1974; 69: 383–393.
34.
RiceJ. Convergence rates for partially spline models. Stat Probab Lett1986; 4: 203–208.
35.
StoneC. Additive regression and other nonparametric models. Ann Stat1985; 13: 689–705.
36.
HeXShiPD. Bivariate tensor-product b-spline in a partly linear model. J Multivariate Anal1996; 58: 162–181.
37.
ZengDCaiJ. Asymptotic results for maximum likelihood estimators in joint analysis of repeated measurements and survival time. Ann Stat2005; 33(5): 2132–2163.
38.
van der VaartAWWellnerJA. Weak convergence and empirical processes, New York: Springer-Verlag, 1996.
39.
ParnerE. Asymptotic theory for the correlated gamma frailty model. Ann Stat1998; 6: 183–214.
