右删失数据下带辅助信息的部分线性分位数回归模型经验似然推断

Empirical likelihood inference of partial linear quantile regression model with auxiliary information under right censored data

利用经验似然结合辅助信息的方法来改进右删失数据下部分线性分位数回归模型的参数估计,通过数值模拟在不同的右删失数据和不同的未知非参数函数下,比较了传统部分线性分位数回归方法和带辅助信息的部分线性分位数回归方法的估计结果的均方误差,进而证明了利用经验似然方法结合辅助信息的部分线性分位数回归方法更有效.

Empirical likelihood combined with auxiliary information was used to improve the parameter estimation of partial linear quantile regression model under right censored data.The mean square error of estimation results of traditional partial linear quantile regression method and partial linear quantile regression method with auxiliary information were compared through numerical simulation under different right censored data and different unknown nonparametric functions.It was proved that the partial linear quantile regression method using empirical likelihood method combined with auxiliary information was more effective.