Proportional hazards model with covariates subject to measurement error.

When covariates of a proportional hazards model are subject to measurement error, the maximum likelihood estimates of regression coefficients based on the partial likelihood are asymptotically biased. Prentice (1982, Biometrika 69, 331-342) presents an example of such bias and suggests a modified partial likelihood. This paper applies the corrected score function method (Nakamura, 1990, Biometrika 77, 127-137) to the proportional hazards model when measurement errors are additive and normally distributed. The result allows a simple correction to the ordinary partial likelihood that yields asymptotically unbiased estimates; the validity of the correction is confirmed via a limited simulation study.

[1]  M. Kendall,et al.  The advanced theory of statistics , 1945 .

[2]  Calyampudi Radhakrishna Rao,et al.  Linear Statistical Inference and its Applications , 1967 .

[3]  David R. Cox,et al.  Regression models and life tables (with discussion , 1972 .

[4]  Calyampudi R. Rao,et al.  Linear Statistical Inference and Its Applications. , 1975 .

[5]  J. Kalbfleisch,et al.  The Statistical Analysis of Failure Time Data , 1980 .

[6]  R. Prentice.  Covariate measurement errors and parameter estimation in a failure time regression model , 1982 .

[7]  W. Dixon,et al.  BMDP statistical software , 1983 .

[8]  Raymond J. Carroll,et al.  Covariate Measurement Error in Logistic Regression , 1985 .

[9]  Mortality of registered A-bomb survivors in Nagasaki, Japan, 1970-1984. , 1985, Radiation research.

[10]  J. T. Hwang.  Multiplicative Errors-in-Variables Models with Applications to Recent Data Released by the U.S. Department of Energy , 1986 .

[11]  Raymond J. Carroll,et al.  Conditional scores and optimal scores for generalized linear measurement-error models , 1987 .

[12]  Eizo Tajima,et al.  US-Japan Joint Reassessment of Atomic Bomb Radiation Dosimetry in Hiroshima and Nagasaki (Final Report) , 1987 .

[13]  L. Stefanski.  Unbiased estimation of a nonlinear function a normal mean with application to measurement err oorf models , 1989 .

[14]  Y. Okumura,et al.  Apparently beneficial effect of low to intermediate doses of A-bomb radiation on human lifespan. , 1990, International journal of radiation biology.

[15]  A S Whittemore,et al.  Censored survival data with misclassified covariates: a case study of breast-cancer mortality. , 1990, Journal of the American Statistical Association.

[16]  Raymond J. Carroll,et al.  Score Tests in Generalized Linear Measurement Error Models , 1990 .

[17]  D. Pierce,et al.  Allowing for random errors in radiation dose estimates for the atomic bomb survivor data. , 1990, Radiation research.