Competing risks data often occurs in survival analysis. The Competing risks arise when an individual is at risk of several mutually exclusive causes, such as multiple causes of death, and the presence of one cause prevents the happening of others (Crowder 2001). There is a little attention given to develop inference methods for the competing risks data where covariates are vulnerable to measurement error. The traditional inferential procedures give biased estimations when covariates are mismeasured. In this paper, we propose an estimation method for producing unbiased estimators based on the proportional cause-specific hazards model featuring covariate measurement error. We employ the likelihood correction technique to fix measurement errors. Derive the asymptotic properties of the estimators of the proposed method. Explore the accomplishments of the proposed method via Monte Carlo simulation studies. Also, we use a real dataset to illustrate the proposed method.
