Modèle de Cox : estimation par sélection de modèle et modèle de chocs bivariés
[Cox model: estimation via model selection and bivariate shock model]
Thèses d'Orsay, no. 583 (2000) , 144 p.

This thesis contains two independent parts, both based on the Cox model.

In the first chapter which presents a joint work with Gwénaëlle Castellan, the Cox model is considered when the regression function of the covariates is not necessarily linear. To estimate this regression function, we devise a nonparametric estimation by model selection. A model is defined as a L -ball of some finite-dimensional linear space of functions. In each model, the regression function is estimated by maximizing the Cox partial log-likelihood. We define then some penalized maximum partial log-likelihood estimator, from this collection of estimators. We give a risk bound for our estimator, in comparison to the smallest risk bound over the considered estimators collection.

In the second chapter, we propose a semiparametric shock model in order to model situations in demography where the biographies of a pair of individuals cannot be considered as independent. For that purpose, we construct two dependent counting processes representing these biographies in such a way that, whenever either one of both counting processes jumps, the hazard rate of the other one is instantaneously multiplied by a constant, called a shock parameter. Moreover, these counting processes may be censored. In such a context, assuming a Cox model, we propose maximum partial log-likelihood estimators for the shock parameters and for the Cox regression parameters, from a sample of independent and identically distributed, possibly censored pairs. Consistency and asymptotic normality of these estimators are established. We illustrate our results with simulations.

Classification: 62G05, 62G07, 62M09, 62F10, 62J15, 62J15, 62P10, 62P25
Keywords: Bivariate censored data, Bivariate survival analysis, Cox model, Kullback-Leibler Information, Model selection, Nonparametric estimation, Penalization.
     author = {Letu\'e, Fr\'ed\'erique},
     title = {Mod\`ele de {Cox} : estimation par s\'election de mod\`ele et mod\`ele de chocs bivari\'es},
     series = {Th\`eses d'Orsay},
     publisher = {Universit\'e de Paris-Sud U.F.R. Scientifique d'Orsay},
     number = {583},
     year = {2000},
     language = {fr},
     url = {}
AU  - Letué, Frédérique
TI  - Modèle de Cox : estimation par sélection de modèle et modèle de chocs bivariés
T3  - Thèses d'Orsay
PY  - 2000
IS  - 583
PB  - Université de Paris-Sud U.F.R. Scientifique d'Orsay
UR  -
LA  - fr
ID  - BJHTUP11_2000__0583__P0_0
ER  - 
%0 Book
%A Letué, Frédérique
%T Modèle de Cox : estimation par sélection de modèle et modèle de chocs bivariés
%S Thèses d'Orsay
%D 2000
%N 583
%I Université de Paris-Sud U.F.R. Scientifique d'Orsay
%G fr
%F BJHTUP11_2000__0583__P0_0
Letué, Frédérique. Modèle de Cox : estimation par sélection de modèle et modèle de chocs bivariés. Thèses d'Orsay, no. 583 (2000), 144 p.

[1] Aalen, O. Nonparametric inference in connection with multiple decrement models. Scand. J. Statist. 3, 1 (1976), 15-27. | MR | Zbl

[2] Aalen, O. Nonparametric inference for a family of counting processes. Ann. Statist. 6, 4 (1978), 701-726. | MR | Zbl

[3] Aalen, O. O., Borgan, Ø., Keiding, N., and Thormann, J. Interaction between life history events. Nonparametric analysis for prospective and retrospective data in the presence of censoring. Scand. J. Statist. 7, 4 (1980), 161-171. | MR | Zbl

[4] Akaike, H. Information theory and an extension of the maximum likelihood principle. 267-281. | MR | Zbl

[5] Andersen, P. K., Borgan, Ø., Gill, R. D., and Keiding, N. Statistical models based on counting processes. Springer-Verlag, New York, 1993. | MR | Zbl | DOI

[6] Andersen, P. K., and Gill, R. D. Cox's regression model for counting processes: a large sample study. Ann. Statist. 10, 4 (1982), 1100-1120. | MR | Zbl | DOI

[7] Bailey, K. R. Asymptotic equivalence between the Cox estimator and the general ML estimators of regression and survival parameters in the Cox model. Ann. Statist. 12, 2 (1984), 730-736. | MR | Zbl | DOI

[8] Barron, A. R., Birgé, L., and Massart, P. Risk bounds for model selection via penalization. Probab. Theory Related Fields 113, 3 (1999), 301-413. | MR | Zbl | DOI

[9] Barron, A. R., AND Sheu, C.-H. Approximation of density functions by sequences of exponential families. Ann. Statist. 19, 3 (1991), 1347-1369. | MR | Zbl | DOI

[10] Bickel, P. J., Klaassen, C. A. J., Ritov, Y., and Wellner, J. A. Efficient and adaptive estimation for semiparametric models. Johns Hopkins University Press, Baltimore, MD, 1993 | MR | Zbl

[11] Birgé, L., and Massart, P. A generalized C p criterion for gaussian model selection. Tech. rep., En préparation.

[12] Birgé, L., and Massart, P. Model selection from a non-asymptotic point of view. En préparation.

[13] Birgé, L., and Massart, P. Rates of convergence for minimum contrast estimators. Probab. Theory Related Fields 97, 1-2 (1993), 113-150. | MR | Zbl | DOI

[14] Birgé, L., and Massart, P. Gaussian model selection. Preprint 2000.05, Université de Paris-Sud, 2000. | MR | Zbl

[15] Birgé, L., and Rozenholc, Y. How many bins should be put in a regular histogram. Technical report, Université Paris VI, 1999. | Zbl | Numdam

[16] Breslow, N. E. Covariance analysis of censored survival data. Biometrics 30 (1974), 89-99. | DOI

[17] Breslow, N. E., and Crowley, J. A large sample study of the life table and product limit estimates under random censorship. Ann. Statist. 2 (1974), 437-453. Collection of articles dedicated to Jerzy Neyman on his 80th birthday. | MR | Zbl | DOI

[18] Bretagnolle, J. Processus ponctuels. Cours de DEA, Université de Paris Sud, 1997-98.

[19] Castellan, G. Modified Akaike's criterion for histogram density estimation. Preprint 99.61, Université de Paris-Sud, 1999.

[20] Castellan, G. Density estimation via exponential model selection. Preprint 2000.25, Université de Paris-Sud, 2000. | MR | Zbl

[21] Castellan, G. Sélection d'histogrammes ou de modèles exponentiels de polynômes par morceaux à l'aide d'un critère de type Akaike. Thèse 6039, Université de Paris-Sud, 2000.

[22] Clayton, D. A model for association in bivariate life tables and its application in epidemiological studies of familial tendency in chronic disease incidence. Biometrika 65, 1 (1978), 141-151. | MR | Zbl | DOI

[23] Clayton, D., and Cuzick, J. Multivariate generalizations of the proportional hazards model. J. Roy. Statist. Soc. Ser. A 148, 2 (1985), 82-117. | MR | Zbl | DOI

[24] Cox, D. D. Multivariate smoothing spline functions. SIAM J. Numer. Anal. 21, 4 (1984), 789-813. | MR | Zbl | DOI

[25] Cox, D. R. Regression models and life-tables. J. Roy. Statist. Soc. Ser. B 34 (1972), 187-220. With discussion by F. Downton, Richard Peto, D. J. Bartholomew, D. V. Lindley, P. W. Glassborow, D. E. Barton, Susannah Howard, B. Benjamin, John J. Gart, L. D. Meshalkin, A. R. Kagan, M. Zelen, R. E. Barlow, Jack Kalbfleisch, R. L. Prentice and Norman Breslow, and a reply by D. R. Cox. | MR | Zbl

[26] Cox, D. R., and Oakes, D. Analysis of survival data. Chapman & Hall, London, 1984. | MR

[27] Dabrowska, D. M. Kaplan-Meier estimate on the plane. Ann. Statist. 16, 4 (1988), 1475-1489. | MR | Zbl | DOI

[28] Dabrowska, D. M. Kaplan-Meier estimate on the plane: weak convergence, LIL, and the bootstrap. J. Multivariate Anal. 29, 2 (1989), 308-325. | MR | Zbl | DOI

[29] Dudley, R. M. The sizes of compact subsets of Hilbert space and continuity of Gaussian processes. J. Functional Analysis 1 (1967), 290-330. | MR | Zbl | DOI

[30] Gill, R. D., Van Der Laan, M. J., and Wellner, J. A. Inefficient estimators of the bivariate survival function for three models. Ann. Inst. H. Poincaré Probab. Statist. 31, 3 (1995), 545-597. | MR | Zbl | Numdam

[31] Gross, A. J., and Clark, V. A. Survival distributions: Reliability applications in the biomedical sciences. Wiley Series in Probability and Mathematical Statistics., New York, 1975. | Zbl

[32] Gross, S. T., and Huber, C. Matched pair experiments: Cox and maximum likelihood estimation. Scand. J. Statist. 14, 1 (1987), 27-41. | MR | Zbl

[33] Hastie, T. J., and Tibshirani, R. J. Generalized additive models. Chapman and Hall Ltd., London, 1990. | MR | Zbl

[34] Hougaard, P. Modelling multivariate survival. Scand. J. Statist. 14, 4 (1987), 291-304. | MR | Zbl

[35] Jacod, J. On the stochastic intensity of a random point process over the half line. Technical report 15, Department of Statistics, Princeton University, 1973.

[36] Jacod, J. Multivariate point processes: predictable projection, Radon-Nikodým derivatives, representation of martingales. Z. Wahrscheinlichkeitstheorie und Verw. Gebiete 31 (1974/75), 235-253. | MR | Zbl | DOI

[37] Johansen, S. An extension of Cox's regression model. Internat. Statist. Rev. 51, 2 (1983), 165-174. | MR | Zbl | DOI

[38] Kalbfleisch, J. D., and Prentice, R. L. The statistical analysis of failure time data. John Wiley and Sons, New York-Chichester-Brisbane, 1980. Wiley Series in Probability and Mathematical Statistics. | MR | Zbl

[39] Kaplan, E. L., and Meier, P. Nonparametric estimation from incomplete observations. J. Amer. Statist. Assoc. 53 (1958), 457-481. | MR | Zbl | DOI

[40] Keiding, N. Statistical inference in the Lexis diagram. Philos. Trans. Roy. Soc. London Ser. A 332, 1627 (1990), 487-509. | MR | Zbl | DOI

[41] Lawless, J. F. Statistical models and methods for lifetime data. John Wiley & Sons Inc., New York, 1982. Wiley Series in Probability and Mathematical Statistics. | MR | Zbl

[42] Lelièvre, E., Bonvalet, C., and Bry, X. Analyse biographique des groupes, les avancées d'une recherche en cours. Population 4 (1997), 803-830. | DOI

[43] Mckeague, I. W., and Utikal, K. J. Inference for a nonlinear counting process regression model. Ann. Statist. 18, 3 (1990), 1172-1187. | MR | Zbl | DOI

[44] Næs, T. The asymptotic distribution of the estimator for the regression parameter in Cox's regression model. Scand. J. Statist. 9, 2 (1982), 107-115. | MR | Zbl

[45] Nelson, W. Hazard plotting for incomplete failure data. J. Qual. Technol. 1 (1969), 27-52. | DOI

[46] Nelson, W. Theory and applications of hazard plotting for censored failure data. Technometrics 14 (1972), 945-965. | DOI

[47] Nielsen, G. G., Gill, R. D., Andersen, P. K., and Sørensen, T. I. A. A counting process approach to maximum likelihood estimation in frailty models. Scand. J. Statist. 19, 1 (1992), 25-43. | MR | Zbl

[48] Oakes, D. Bivariate survival models induced by frailties. J. Amer. Statist. Assoc. 84, 406 (1989), 487-493. | MR | Zbl | DOI

[49] O'Sullivan, F. Nonparametric estimation in the Cox model. Ann. Statist. 21, 1 (1993), 124-145. | MR | Zbl

[50] Pons, O. A test of independence between two censored survival times. Scand. J. Statist. 13, 3 (1986), 173-185. | MR | Zbl

[51] Pons, O., and De Turckheim, È. Tests of independence for bivariate censored data based on the empirical joint hazard function. Scand. J. Statist. 18, 1 (1991), 21-37. | MR | Zbl

[52] Rebolledo, R. La méthode des martingales appliquée à l'étude de la convergence en loi de processus. Bull. Soc. Math. France Mém., 62 (1979), v+125 pp. (1980). | MR | Zbl | Numdam

[53] Senoussi, R. Problème d'identification dans le modèle de Cox. Ann. Inst. H. Poincaré Probab. Statist. 26, 1 (1990), 45-64. | MR | Zbl | Numdam

[54] Stone, C. J. Optimal global rates of convergence for nonparametric regression. Ann. Statist. 10, 4 (1982), 1040-1053. | MR | Zbl

[55] Tsiatis, A. A. A large sample study of Cox's regression model. Ann. Statist. 9, 1 (1981), 93-108. | MR | Zbl

[56] Van Der Laan, M. J. Efficient estimation in the bivariate censoring model and repairing NPMLE. Ann. Statist. 24, 2 (1996), 596-627. | MR | Zbl

[57] Van Der Vaart, A. W. Asymptotic statistics. Cambridge University Press, Cambridge, 1998. | MR | Zbl

[58] Van Der Vaart, A. W. Semiparametric statistics. Cours, École d'Été de Probabilités de Saint-Flour XXIX, 1999. | Zbl

[59] Van Der Vaart, A. W., and Wellner, J. A. Weak convergence and empirical processes. Springer-Verlag, New York, 1996. With applications to statistics. | MR | Zbl | DOI

[60] Vaupel, J. W., Manton, K. G., and Stallard, E. The impact of heterogeneity in individual frailty on the dynamics of mortality. Demography, 16 (1979), 439-454. | DOI

[61] Wahba, G. Spline models for observational data. Society for Industrial and Applied Mathematics (SIAM), Philadelphia, PA, 1990. | MR | Zbl