arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1404.7084stat.ME
收藏

基于Bernstein型多项式的高效稳健密度估计

Efficient and Robust Density Estimation Using Bernstein Type Polynomials

Zhong Guan

摘要:

提出了一种利用Bernstein多项式对未知底层分布进行参数化和光滑化的方法,并对该方法进行了验证和研究。该模型可以近似任何具有有界和足够光滑密度的分布。近似模型结果是β分布β $(i+1, m-i+1)$、$i=0,\ldots, m$的混合,对于某个最佳程度$m$。给出了一种选择Bernstein多项式最佳次数$m$的简单变点估计方法。在一定条件下,给出了在$L_2$距离上以几乎参数速率一致的极大似然密度估计.仿真研究表明,该方法可以同时获得平滑性和准确性。该模型还可用于估计未知分布的一些函数,如总体均值。作为说明,所提出的方法被应用到三个不同类型的数据集,包括一个微阵列数据集。

英文摘要:

Method of parameterizing and smoothing the unknown underling distributions using Bernstein polynomials is proposed, verified and investigated. Any distribution with bounded and smooth enough density can be approximated by the proposed model. The approximating model turns out to be a mixture of beta distributions beta$(i+1, m-i+1)$, $i=0,\ldots, m$, for some optimal degree $m$. A simple change-point estimating method for choosing optimal degree $m$ of the Bernstein polynomials is also presented. The proposed methods give maximum likelihood density estimate which is consistent in $L_2$ distance at an almost parametric rate under some conditions. Simulation study shows that one can benefit from both the smoothness and the accuracy by using the proposed method. The proposed model can also be used to estimate some functional of the unknown distribution such as population mean. As illustration, the proposed methods are applied to three different type data sets including a microarray data set.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑