Abstract
In this paper the problem of comparing k density functions from survival data is considered. Two non-parametric tests based on (two different) generalizations of the L 1 measure are adapted to the censored context. The asymptotic distribution of the test statistics is derived, and an approximation based on resampling methods is proposed. The relative power of the tests is investigated through a Monte Carlo simulation study. Results suggest that the tests exhibit good power when no one of the survival functions dominates the others, especially when the censoring distribution is the same along the k groups and the censoring percentage is small.
Similar content being viewed by others
References
Abrams DI, Goldman AI, Launer C, Korvick JA, Neaton JD, Crane LR, Grodesky M, Wakefield S, Muth K, Kornegay S, Cohn DJ, Haris A, Luskin-Hawk R, Markowitz N, Sampson JH, Thompson M, Deyton L, The Terry Beirn Community Programs for Clinical Research on AIDS (1994) Comparative trial of didanosine and zalcitabine in patients with human immunodeficiency virus infection who are intolerant or have failed zidovudine therapy, N Engl J Med 330:657–662
Beran R (1982) Estimated sampling distributions: the bootstrap and competitors. Ann Stat 9: 1196–1217
Blum JR, Susarla V (1980) Maximal deviation theory of density and failure rate function estimates based on censored data. J Multivar Anal 5: 213–222
Cao R, Van Keilegom I (2006) Empirical likelihood tests for two-sample problems via nonparametric density estimation. Can J Stat 34: 61–77
Efron B (1981) Censored data and the bootstrap. J Am Stat Assoc 76(374): 312–319
Efron B (1982) The jacknife, the bootstrap and other resampling plans. Society for Industrial and Applied Mathematics, Philadelphia
Földes A, Retjö L, Winter BB (1981) Strong consistency properties of nonparametric estimators for randomly censored data, II: estimation for density and failure rate. Period Math Hung 12: 15–29
Gonzalez-Manteiga W, Cao R, Marron JS (1996) Bootstrap selection of the smoothing parameter in nonparametric hazard rate estimation. J Am Stat Assoc 91(435): 1130–1140
Harrington DP (2005) Linear rank tests in survival analysis. Enciclopedia of Biostatistics. Wiley Interscience, New York
Horváth L (1991) On L p -norms of multivariate density estimators. Ann Stat 19(4): 1933–1949
Kaplan EL, Meier P (1958) Non parametric estimation from incomplete observations. J Am Stat Assoc 53: 457–481
Marron JS, Padgett WJ (1987) Asymptotically optimal bandwidth selection for kernel density estimators from randomly right-censored samples. Ann Stat 15(4): 1520–1535
Martínez-Camblor P (2010a) Comparing k-independent and right censored samples based on the likelihood ratio. Comput Stat 25: 363–374
Martínez-Camblor P (2010b) Nonparametric k-sample test based on kernel density estimator for paired design. Comput Stat Data Anal 54: 2035–2045
Martínez-Camblor P (2011) Testing the equality among distribution functions from independent and right censored samples via Cramér-von Mises criterion. J Appl Stat 38(6): 1117–1131
Martínez-Camblor P, de Uña-Álvarez J (2009a) Non parametric k-sample tests: density functions versus distribution functions. Comput Stat Data Anal 53(9): 3344–3357
Martínez-Camblor P, de Uña-Álvarez J (2009b) Studying the bandwidth in k-sample smooth tests. Discussion Papers in Statistics and Operation Research. Universidade de Vigo, Dpto. de Estadística e Investigación Operativa. Report 09/02
Martínez-Camblor P, de Uña-Álvarez J, Corral N (2008) k-sample test based on the common area of kernel density estimators. J Stat Plan Inference 138(12): 4006–4020
Martínez-Camblor P, Larrañaga N, Sarasqueta C, Basterretxea M (2009a) Esa corporeidad mortal y rosa. Análisis del tiempo libre de enfermedad del cáncer de mama en Gipuzkoa en presencia de riesgos competitivos. Gaceta Sanitaria 23(6): 554–557
Mielniczuk J (1986) Some asymptotic properties of kernel estimators of density function in case of censored data. Ann Stat 14: 766–773
Zhang B (1996) Some asymptotic result for kernel density estimation under random censorship. Benoulli 2(2): 183–198
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Martínez-Camblor, P., de Uña-Álvarez, J. Density comparison for independent and right censored samples via kernel smoothing. Comput Stat 28, 269–288 (2013). https://doi.org/10.1007/s00180-011-0298-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00180-011-0298-5