Progressive censoring is essential for researchers in industry as a mean to remove subjects before the final termination point in order to save time and reduce cost. Recently, kernel density estimation has been intensively investigated due to its asymptotic properties and applications. In this paper, we investigate the asymptotic properties of the kernel density estimators based on progressive type-II censoring and their application to hazard function estimation. A bias-adjusted kernel density estimator is also proposed. Our simulation indicates that the kernel density estimates under progressive type-II censoring is competitive compared with kernel density estimates under simple random sampling, depending on the censoring schemes. An example regarding failure times of aircraft windshields is used to illustrate the proposed methods.

中文翻译：

---

基于渐进式Ⅱ类删失的核密度估计

渐进式审查对于行业研究人员至关重要，它是在最终终止点之前删除主题的一种手段，以节省时间并降低成本。近来，由于其渐近性质和应用，对内核密度估计进行了深入研究。在本文中，我们研究了基于渐进式II型删失的核密度估计量的渐近性质及其在危险函数估计中的应用。还提出了偏差调整后的核密度估计器。我们的仿真表明，根据检查方案的不同，渐进式II型审查下的核密度估计与简单随机抽样下的核密度估计相比具有竞争力。以飞机挡风玻璃故障时间为例，说明了所提出的方法。