arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1503.06370stat.ME
收藏

基于$κ$ - $G$先验的线性回归贝叶斯变量选择

Bayesian Variable Selection for Linear Regression with the $κ$-$G$ Priors

Zichen Ma, Ernest Fokoué

摘要:

在本文中,我们介绍了一种新的方法,贝叶斯变量选择的线性回归,是独立于传统的指标方法。对角矩阵$\mathbf{G}$被引入到系数向量$\boldsymbolβ$的先验中,其中在对角线上的限定在$0$和$1$之间的每个$g_j$用作对应的$β_j$的稳定器。在数学上,有希望的变量具有接近于$0$的$g_j$值,而对应于无希望的变量的$g_j$的值接近于$1$。本文在正交性下证明了这一性质以及其它渐近性质。在计算上,每个$g_j$的样本路径通过Metropolis-within- Gibbs抽样方法获得。此外,在本文中,我们给出了两个模拟来验证这种方法的能力,在变量选择。

英文摘要:

In this paper, we introduce a new methodology for Bayesian variable selection in linear regression that is independent of the traditional indicator method. A diagonal matrix $\mathbf{G}$ is introduced to the prior of the coefficient vector $\boldsymbolβ$, with each of the $g_j$'s, bounded between $0$ and $1$, on the diagonal serves as a stabilizer of the corresponding $β_j$. Mathematically, a promising variable has a $g_j$ value that is close to $0$, whereas the value of $g_j$ corresponding to an unpromising variable is close to $1$. This property is proven in this paper under orthogonality together with other asymptotic properties. Computationally, the sample path of each $g_j$ is obtained through Metropolis-within-Gibbs sampling method. Also, in this paper we give two simulations to verify the capability of this methodology in variable selection.

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。

↑