Abstract
The local behavior of oscillation modulus of the product-limit (PL) process and the cumulative hazard process is investigated when the data are subjected to random censoring. Laws of the iterated logarithm of local oscillation modulus for the PL-process and the cumulative hazard process are established. Some of these results are applied to obtain the almost sure best rates of convergence for various types of density estimators as well as the Bahadur-Kiefer type process.
Similar content being viewed by others
References
Kaplan, E. L., Meier, P., Nonparametric estimation from incomplete observations,J. Amer. Statist. Assoc., 1958, 53: 457.
Diehl, S., Stute, W., Kernel density and hazard function in presence of censoring,J. Multivariate Anal., 1988, 25: 299.
Csörgö, M., Horváth, L., The rate of strong uniform consistency for the product-limit estimator,Z. Wahrsch. Verw. Gebiete, 1983, 62: 411.
Major, P., Rejtö, L., Strong embedding of the estimator of the distribution function under random censorship,Ann. Statist., 1988, 16(3): 1113.
Shorack, G. R., Wellner, J. A.,Empirical Processes With Applications to Statistics, New York: John Wiley, 1986.
Stute, W., The oscillation behavior of empirical processes,Ann. Probab., 1982, 10(1): 86.
Mason, D. M., Shorack, G., Wellner, J. A., Strong limit theorems for oscillation modulus of the empirical process,Z. Wahrsch. Verw. Gebiete, 1983, 65(1): 83.
Zhou, Y., The rate convergence of oscillation modulus for product-limit processes,Chinese Science Bulletin, 1995, 40: 1409.
Ralescu, S. S., The asymptotic law of the local oscillation modulus of the empirical process,J. Statist. Plan. Inference, 1993, 35(1): 139.
Lo, S. H., Mack, Y. P., Wang, J. T., Density and hazard rate estimation for censored data via strong representations of the Kaplan-Meier estimator,Probab. Th. Rel. Fields, 1989, 80: 461.
Aly, A. A., Csörgo, M., Horváth, L., Strong approximation of the quantile process of the product-limit estimator,J. Multivariate Anal., 1985, 16: 185.
Burke, M. D., Csörgo, S., Horváth L. Strong approximations of some biometric estimate under random censorship,Z. Wahrsch. Verw. Gebiete, 1981, 56(1): 87.
Blum, J. R., Susarla, V., Maximal deviation theory of density and failure estimates based on censored data, inMultivariate Analysis V (ed. Krishnaiah, P. R.), Amsterdam: North-Holland, 1980, 213.
Zhang, B., A law of the iterated logarithm for kernel estimator of hazard functions under random censorship,Scand. J. Statist. Appl., 1996, 23(1): 37.
Stute, W., A law of the logarithm for kernel density estimators,Ann. Probab., 1982, 10: 414.
Mielniczuk, J., Some asymptotic properties of kernel estimators of a density function in case of censored data,Ann. Statist., 1986, 14: 766.
Einmahl, J. H. J., Ruymgaart, F. H., The almost sure behavior of the oscillation modulus of the multivariate empirical process,Statist. Probab. Lett., 1987, 6: 87.
Mallows, C. L., An inequality involving multinomial probabilities,Biometrika, 1968, 55: 422.
Alexander, K. S., Rates of growth and sample modulus for weighted empirical process indexed by sets,Probab. Th. Fields, 1987, 75(3): 379.
Author information
Authors and Affiliations
Additional information
Project supported in part by the National Natural Science Foundation of China (Grant No. 19701037).
Rights and permissions
About this article
Cite this article
Yong, Z., Liuquan, S. & Yip, P.S.F. The almost sure behavior of the oscillation modulus for PL-process and cumulative hazard process under random censorship. Sci. China Ser. A-Math. 42, 225–237 (1999). https://doi.org/10.1007/BF02879056
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF02879056