Functional regression via variational Bayes.

Functional regression via variational Bayes.
复制标题

DOI:
10.1214/11-ejs619
复制
发表时间:
2011-01-01
影响因子:
1.1
通讯作者:
Crainiceanu C
Crainiceanu C
中科院分区:
数学3区
文献类型:
--
作者:
Goldsmith J;Wand MP;Crainiceanu C

文献摘要

参考文献

被引文献

相似文献

我们引入变分贝叶斯方法,用于函数回归分析中的快速近似推理。标准横截面和日益常见的纵向设置均得到处理。该方法允许进行贝叶斯函数回归分析,而无需蒙特卡罗方法的计算开销。使用近似变分方法和簇的非参数重采样获得模型参数的置信区间。后一种方法是可能的,因为我们的变分贝叶斯函数回归方法计算效率高。模拟研究表明,变分贝叶斯在估计感兴趣的参数和逼近模型参数的马尔可夫链蒙特卡罗采样联合后验分布方面具有很高的准确性。这些方法普遍适用,但其动机是对多发性硬化症患者进行纵向神经影像学研究。模拟中使用的代码作为网络补充提供。
We introduce variational Bayes methods for fast approximate inference in functional regression analysis. Both the standard cross-sectional and the increasingly common longitudinal settings are treated. The methodology allows Bayesian functional regression analyses to be conducted without the computational overhead of Monte Carlo methods. Confidence intervals of the model parameters are obtained both using the approximate variational approach and nonparametric resampling of clusters. The latter approach is possible because our variational Bayes functional regression approach is computationally efficient. A simulation study indicates that variational Bayes is highly accurate in estimating the parameters of interest and in approximating the Markov chain Monte Carlo-sampled joint posterior distribution of the model parameters. The methods apply generally, but are motivated by a longitudinal neuroimaging study of multiple sclerosis patients. Code used in simulations is made available as a web-supplement.
DOI: 10.1214/aoms/1177729694
发表时间: 1951-01-01
影响因子: --
作者:
KULLBACK, S;LEIBLER, RA
通讯作者: LEIBLER, RA
DOI: 10.1136/jnnp.2006.112177
发表时间: 2008-04-01
影响因子: 11
作者:
Lin, X.;Tench, C. R.;Constantinescu, C. S.
通讯作者: Constantinescu, C. S.
DOI: 10.1016/s0167-7152(99)00036-x
发表时间: 1999-10-15
影响因子: 0.8
作者:
Cardot, H;Ferraty, F;Sarda, P
通讯作者: Sarda, P
DOI: 10.1198/016214507000000527
发表时间: 2007-09-01
影响因子: 3.7
作者:
Reiss, Philip T.;Ogden, R. Todd
通讯作者: Ogden, R. Todd
DOI: 10.1214/088342304000000026
发表时间: 2004-02-01
影响因子: 5.7
作者:
Jordan, MI
通讯作者: Jordan, MI