部分线性回归模型的加权似然推断
WEIGHTED LIKELIHOOD INFERENCE FOR PARTIALLY LINEAR REGRESSION MODELS
针对不同来源的几组相关数据集, 研究了部分线性模型的加权似然推断问题, 给出了加权似然 估计的相合性和渐近正态性. 模拟结果表明在均方误差意义下, 加权似然得到的估计优于经典的极大似然估计, 并把新的估计方法应用到艾滋病临床试验数据分析中.
This article studies weighted likelihood inference for partially linear models when several related data sets are available from different sources. The estimator derived from the weighted likelihood can reduce the mean squared error (MSE) of the classical maximum likelihood estimator. The improvement is illustrated numerically. The resulting estimator is shown to be consistent and asymptotically normal. A real data set from an AIDS clinical trial is also analyzed.
有效性 / 预光滑 / 半参数估计 / 加权似然. {{custom_keyword}} /
/
〈 | 〉 |