Empirical likelihood inference for the odds ratio of two survival functions under right censoring

Abstract

a b s t r a c tThis paper develops a new method for estimating survival odds ratios under right censoring using the censored empirical likelihood. Results of a simulation study show that under small sample sizes the proposed method outperforms the corresponding Wald-type CI and the log-transformed Wald-type CI.

Novel empirical likelihood inference for the mean difference with right-censored data

“…This fact shows the consistency of the IID-EL over the EL-WW. Also, the EL-WW attains the nominal levels with samples (m, n) = (25, 30), (m, n) = (30,25) or (m, n) = (60, 60), while the IID-EL reaches them quickly even with sample size (m, n) = (10,15) or (m, n) = (15, 10). 2.…”

“…Based on the data set, the EL-WW-based, IID-EL-based, IID-AEL-based, and IID-MEL-based confidence intervals and their average lengths are calculated according to Theorem 1, Wang and Wang 19 ), Theorems 2-4 in Section for 90% and 95% confidence levels. For each fixed value of c 0 , c 1 , λ 1 , λ 2 and λ 3 and sample sizes (m, n) = (10, 15), (15,10), (25,30), (30,25), and (60, 60), the process is repeated 10, 000 times. Censoring rates of 10%, 25% and 40% have been considered.…”

“…Using smoothed EL methods, Zhao and Zhao 24 constructed confidence intervals for the ratio and difference of two hazard functions. The EL is also used by Zhao et al 25 to estimate odds ratios of survival functions. To our knowledge, no paper used the independent and identically distributed (i.i.d.)…”

On Kernel-Based Estimator of Odds Ratio Using Different Stratified Sampling Schemes

“…[1] presented some properties of OR based on the categorical data. Some of the recent contributions are by [19], [13], [39], [37], [18], [26].…”