Empirical likelihood confidence intervals for the differences of quantiles with missing data

Suppose there are two nonparametric populations X and Y with missing data on both of them. Random imputation is used to fill in missing data, so the “complete”samples of X and Y can be constructed. Then the empirical likelihood confidence intervals for the differences of quantile are constructed. Key words：Empirical likelihood; Confidence interval; Quantile; Missing data; Imputation. 2000 MR subject classification: 62G05, 62E20