Abstract
This paper considers estimation under the Cox proportional hazards model with right-censored event times in the presence of covariates missing not at random (MNAR). We propose an approach derived from likelihood estimation utilizing supplementary information. We show that available additional information not only helps to account appropriately for the missing covariates but also leads to estimation procedures which are natural and easy to implement. A medical example is used throughout the paper to motivate the problem and to illustrate the proposed methodology.
Similar content being viewed by others
References
Andersen PK, Borgan O, Gill RD, Keiding N (1992) Statistical models based on counting processes. Springer, New York
Bang H, Robins JM (2005) Doubly robust estimation in missing data and causal inference models. Biometrics 61:962–973
Breslow NE (1972) Discussion of the paper by DR Cox. J R Stat Soc B 34:216–217
Canadian Tuberculosis Standards 6th edn (2007) Edited by Long, R Canadian Lung Association, Canadian Thoracic Society and Tuberculosis Prevention and Control Centre for Infectious Disease Prevention and Control, Health Canada
Chen MH, Ibrahim JG, Shao QM (2009) Maximum likelihood inference for the Cox regression model with applications to missing covariates. J Multivar Anal 100:2018–2030
Chen HY, Little RJA (1999) Proportional hazards regression with missing covariates. J Am Stat Assoc 94:896–908
Cook VJ, Hernandex-Garduno E, Hu XJ, Elwood RK, FitzGerald JM (2005) The influence of cluster-status of source cases on contact evaluation and the development of secondary active tuberculosis. Proceedings of the American Thoracic Society (PATS) 2 Abstract Issue
Cox DR (1972) Regression models and life tables (with discussion). J R Stat Soc B 34:187–220
Dempster AP, Laird NM, Rubin DB (1977) Maximum likelihood from incomplete data via the EM algorithm. J R Stat Soc B 39:1–22
Efron B, Tibshirani RJ (1993) An introduction to the bootstrap. Monographs on statistics and applied probability, vol 57. Chapman & Hall/CRC, London/Boca Raton
Herring AH, Ibrahim JG (2001) Likelihood-based methods for missing covariates in the Cox proportional hazards model. J Am Stat Assoc 96:292–302
HIV/AIDS Epi Updates November 2007. Centre for Infectious Disease Prevention and Control, Public Health Agency of Canada
Hu XJ, Lawless JF (1996) Estimation from truncated lifetime data with supplementary information on covariates and censoring times. Biometrika 83:747–61
Hu XJ, Lawless JF (1997) Pseudo-likelihood estimation in a class of problems with response-related missing covariates. Can J Stat 25:125–42
Li Y, Dicker L, Zhao S (2010) A new class of Dantzig selectors for censored linear regression models. Harvard University Biostatistics Working Paper Series (March 2, 2010)
Lin DY, Ying Z (1993) Cox regression with incomplete covariate measurements. J Am Stat Assoc 88:1341–1349
Little RJA, Rubin DB (2002) Statistical analysis with missing data. Wiley series in probability and statistics. Wiley, New York
Nan B, Kalbfleisch JD, Yu M (2009) Asymptotic theory for the semiparametric accelerated failure time model with missing data. Ann Stat 37:2351–2376
Paik MC, Tsai WY (1997) On using the Cox proportional hazards model with missing covariates. Biometrika 84:579–593
Qi L, Wang CY, Prentice RL (2005) Weighted estimators for proportional hazards regression with missing covariates. J Am Stat Assoc 100:1250–1263
Reichler MR, Bur S, Reves R, Mangura B, Thompson V, Ford J, Castro KG (2003) Results of testing for human immunodeficiency virus infection among recent contacts of infectious tuberculosis cases in the United States. Int J Tuberc Lung Dis 7(12):5471–5478
Robins JM, Rotnitzky A, Zhao LP (1994) Estimation of regression coefficients when some regressors are not always observed. J Am Stat Assoc 89:846–866
Scharfstein DO, Rotnitzky A, Robins JM (1999) Adjusting for nonignorable drop-out using semiparametric nonresponse models. J Am Stat Assoc 94:1096–1120
Suzuki K (1985) Estimation method of lifetime based on the record of failures during the warranty period. J Am Stat Assoc 80:68–72
Tsiatis AA (2006) Semiparametric theory and missing data. Springer series in statistics. Springer, Berlin
Zeng D, Cai J (2005) Asymptotic results for maximum likelihood estimators in joint analysis of repeated measurements and survival time. Ann Stat 33:2132–2163
Zhou H, Pepe MS (1995) Auxiliary covariate data in failure time regression. Biometrika 82:139–149
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Cook, V.J., Hu, X.J. & Swartz, T.B. Cox Regression with Covariates Missing Not at Random. Stat Biosci 3, 208–222 (2011). https://doi.org/10.1007/s12561-010-9031-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12561-010-9031-0