二值响应数据贝叶斯中位数估计

Bayesian Median Regression of Binary Response Data

中位数回归是一种稳健的估计方法,在实践中有着广泛应用.基于贝叶斯方法研究二值响应数据的中位数估计问题,通过引入合适的潜在变量得到了贝叶斯层次模型,进而得到易于后验抽样的吉布斯抽样程序.为验证新方法估计的稳健性,通过大量数据模拟,并与已有方法进行比较,得到了满意的结果.最后实例数据分析进一步证明了所提方法的有效性.

Median regression is a robust estimation method, which is widely used in practice. In this paper, the median estimation of binary response data is studied based on Bayesian method. The Bayesian hierarchical model is obtained by introducing suitable latent variables, and then the Gibbs sampling program is easy to be obtained. To examine the robustness of the new estimator, some simulation studies were conducted and the result was satisfactory compared with existing method. Finally, the real data analysis was further used to confirm its effectiveness.