Empirical Likelihood Ratio in Terms of Cumulative Hazard Function for Censored Data

It has been shown that (with complete data) empirical likelihood ratios can be used to form confidence intervals and test hypotheses about a linear functional of the distribution function just like the parametric case. We study here the empirical likelihood ratios for right censored data and with parameters that are linear functionals of the cumulative hazard function. Martingale techniques make the asymptotic analysis easier, even for random weighting functions. It is shown that the empirical likelihood ratio in this setting can be easily obtained by solving a one parameter monotone equation.