Biometrika Advance Access originally published online on August 5, 2007
Biometrika 2007 94(3):705-718; doi:10.1093/biomet/asm057
Copyright © 2007 Biometrika Trust
Articles |
Estimation of the mean function with panel count data using monotone polynomial splines
Department of Biostatistics, The University of Iowa, 200 Hawkins Drive, C22 GH, Iowa City, Iowa 52242, U.S.A.
Department of Statistics and Actuarial Science, The University of Iowa, 221 Schaeffer Hall, Iowa City, Iowa 52242, U.S.A.
minggen-lu{at}uiowa.edu
ying-j-zhang{at}uiowa.edu
jian-huang{at}uiowa.edu
Received for publication 1 April 2006. Revision received 1 January 2007.
We study nonparametric likelihood-based estimators of the mean function of counting processes with panel count data using monotone polynomial splines. The generalized Rosen algorithm, proposed by Zhang & Jamshidian (2004), is used to compute the estimators. We show that the proposed spline likelihood-based estimators are consistent and that their rate of convergence can be faster than n1/3. Simulation studies with moderate samples show that the estimators have smaller variances and mean squared errors than their alternatives proposed by Wellner & Zhang (2000). A real example from a bladder tumour clinical trial is used to illustrate this method.
Key Words: Counting process • Empirical process • Isotonic regression • Maximum likelihood estimator • Maximum pseudolikelihood estimator • Monotone polynomial spline • Monte Carlo
References
-
Anderson J. A., Senthilselvan A. Smooth estimates for the hazard function. J. R. Statist. Soc. (1980) B 42:322–7.
Andrews D. F., Herzberg A. M. Data; A Collection of Problems from Many Fields for the Student and Research Worker (1985) New York: Spring-Verlag.
De Boor C. A Practical Guide to Splines (2001) New York: Spring-Verlag.
Byar D. P. The Veterans administration study of chemoprophylaxis for recurrent stage I bladder tumours: comparisons of placebo, pyridoxine, and topical thiotepa. In: Bladder Tumors and Other Topics in Urological Oncology—Macaluso M., Smith P. H., Edsmyr F., eds. (1980) New York: Plenum. 363–70.
Byar D. P., Blackard C., THe Veterans Administration Cooperative Urological Research Group. Comparisons of placebo, pyridoxine, and topical thiotepa in preventing recurrence of stage I bladder cancer. Urology (1977) 10:556–61.[CrossRef][Medline]
Cai T., Betensky R. A. Hazard regression for interval-censored data with penalized spline. Biometrics (2003) 59:570–9.[CrossRef][ISI][Medline]
Huang J. Efficient estimation of the partly linear additive Cox model. Ann. Statist. (1999) 27:1536–63.[CrossRef]
Jamshidian M. On algorithms for restricted maximum likelihood estimation. Comp. Statist. Data Anal. (2004) 45:137–57.[CrossRef]
Jongbloed G. The iterative convex minorant algorithm for nonparametric estimation. J. Comp. Graph. Statist. (1998) 7:310–21.[CrossRef]
Kalbfleisch J. D., Lawless J. F. The analysis of panel count data under a Markov assumption. J. Am. Statist. Assoc. (1985) 80:863–71.[CrossRef][ISI]
Kooperberg C., Stone C. J., Truong Y. K. Hazard regression. J. Am. Statist. Assoc. (1995) 90:78–94.[CrossRef][ISI]
Lee E. W., Kim M. Y. The analysis of correlated panel count data using a continuous-time Markov model. Biometrics (1998) 54:1638–44.[CrossRef][ISI][Medline]
Ramsay J. O. Monotone regression splines in action. Statist. Sci. (1988) 3:425–41.
Rosen J. B. The gradient projection method for nonlinear programming. J. SIAM (1960) 8:181–217.
Rosenberg P. S. Hazard function estimation using B-splines. Biometrics (1995) 51:874–87.[CrossRef][ISI][Medline]
Schumaker L. Spline Functions: Basic Theory (1981) New York: Wiley.
Shen X., Wong W. H. Convergence rate of sieve estimates. Ann. Statist. (1994) 22:580–615.
Stone C. J. The dimensionality reduction principle for generalized additive models. Ann. Statist. (1986) 14:590–606.
Sun J., Kalbfleisch J. D. Estimation of the mean function of point processes based on panel count data. Statist. Sinica. (1995) 5:279–90.
Sun J., Wei L. J. Regression analysis of panel count data with covariate-dependent observation and censoring times. J. R. Statist. Soc. (2000) B 62:293–302.[CrossRef]
Thall P. F. Mixed Poisson likelihood regression models for longitudinal interval count data. Biometrics (1988) 44:197–209.[CrossRef][ISI][Medline]
Thall P. F., Lachin J. M. Analysis of recurrent events: nonparametric methods for random-interval count data. J. Am. Statist. Assoc. (1988) 83:339–47.[CrossRef][ISI]
Van der Vaart A. W., Wellner J. A. Weak Convergence and Empirical Processes with Application to Statistics (1996) New York: Springer-Verlag.
Wei L. J., Lin D. Y., Weissfeld L. Regression analysis of multivariate incomplete failure time data by modeling marginal distributions. J. Am. Statist. Assoc. (1989) 84:1065–73.[CrossRef][ISI]
Wellner J. A., Zhang Y. Two estimators of the mean of a counting process with panel count data. Ann. Statist. (2000) 28:779–814.[CrossRef]
Whittemore A. S., Keller J. B. Survival estimation using splines. Biometrics (1986) 42:495–506.[CrossRef][ISI][Medline]
Zhang Y. A semiparametric pseudolikelihood estimation method for panel count data. Biometrika (2002) 89:39–48.
Zhang Y., Jamshidian M. On algorithms for NPMLE of the failure function with censored data. J. Comp. Graph. Statist. (2004) 13:123–40.[CrossRef]
Zhou S., Shen X., Wolfe D. A. Local asymptotics for regression splines and confidence regions. Ann. Statist. (1998) 26:1760–82.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||