姜云卢, 袁晶晶. 基于MM-算法的局部常数核加权分位数回归估计及应用[J]. 华南师范大学学报(自然科学版), 2013, 45(4).
引用本文: 姜云卢, 袁晶晶. 基于MM-算法的局部常数核加权分位数回归估计及应用[J]. 华南师范大学学报(自然科学版), 2013, 45(4).
Local constant kernel-weighted quantile regression estimation via an MM-algorithm and its applications[J]. Journal of South China Normal University (Natural Science Edition), 2013, 45(4).
Citation: Local constant kernel-weighted quantile regression estimation via an MM-algorithm and its applications[J]. Journal of South China Normal University (Natural Science Edition), 2013, 45(4).

基于MM-算法的局部常数核加权分位数回归估计及应用

Local constant kernel-weighted quantile regression estimation via an MM-algorithm and its applications

  • 摘要: 提出了一种新的基于MM-算法的局部常数核加权计算法,使用该方法得到的分位数回归估计曲线在比较弱的条件下是连续光滑的.讨论了核方法中协调参数的选取问题.最后,数值模拟和实证研究的结果表明,利用所提出的计算方法所得到的非参数曲线估计在重尾误差的情况下具有一定的稳健性.

     

    Abstract: A new local constant kernel weighted computational method based on MM-algorithm is proposed in this paper. The proposed method makes that the calculated quantile regression estimation curve is continuous and smooth under some mild conditions. The selection problem of the tuning parameter is also discussed. Finally, simulation studies and a real data analysis illustrate that the quantile regression estimation curve based on our proposed computational method is robust under the model with heavy-tailed error.

     

/

返回文章
返回