Comparing and weighting imperfect models using D-probabilities.

Comparing and weighting imperfect models using D-probabilities.
复制标题

使用D概率比较和加权不完美模型。

DOI:
10.1080/01621459.2019.1611140
复制
发表时间:
2020
影响因子:
3.7
通讯作者:
Dunson DB
Dunson DB
中科院分区:
数学1区
文献类型:
--
作者:
Li M;Dunson DB

文献摘要

参考文献

被引文献

相似文献

我们提出了一种新的方法分配权重的模型,使用基于分歧的方法(D-概率),依赖于评估参数模型相对于非参数贝叶斯参考使用Kullback-Leibler分歧。D-概率在拟合优度评估、比较不完美模型以及提供模型聚合中使用的模型权重方面非常有用。D-概率避免了贝叶斯模型概率的一些缺点,例如对先验选择的高度敏感性,并且倾向于在更大的模型多样性上放置更高的权重。在一个应用程序的线性模型选择对高斯过程的参考,我们提供了简单的分析形式,例行实施和D-概率自动惩罚模型的复杂性。一些渐近性质的描述,我们提供了有趣的概率解释建议的模型权重。该框架是通过模拟的例子和臭氧数据的应用程序。
We propose a new approach for assigning weights to models using a divergence-based method (D-probabilities), relying on evaluating parametric models relative to a nonparametric Bayesian reference using Kullback-Leibler divergence. D-probabilities are useful in goodness-of-fit assessments, in comparing imperfect models, and in providing model weights to be used in model aggregation. D-probabilities avoid some of the disadvantages of Bayesian model probabilities, such as large sensitivity to prior choice, and tend to place higher weight on a greater diversity of models. In an application to linear model selection against a Gaussian process reference, we provide simple analytic forms for routine implementation and show that D-probabilities automatically penalize model complexity. Some asymptotic properties are described, and we provide interesting probabilistic interpretations of the proposed model weights. The framework is illustrated through simulation examples and an ozone data application.
DOI: 10.2307/2286745
发表时间: 1979-01-01
影响因子: 3.7
作者:
GEISSER, S;EDDY, WF
通讯作者: EDDY, WF
DOI: 10.1109/tit.2018.2805844
发表时间: 2018-04-01
影响因子: 2.5
作者:
Bu, Yuheng;Zou, Shaofeng;Veeravalli, Venugopal V.
通讯作者: Veeravalli, Venugopal V.
DOI: 10.1214/aos/1021379858
发表时间: 2002-04-01
影响因子: 4.5
作者:
Hoffmann, M;Lepski, O
通讯作者: Lepski, O
DOI: 10.1109/tac.1974.1100705
发表时间: 1974-01-01
影响因子: 6.8
作者:
AKAIKE, H
通讯作者: AKAIKE, H
DOI: 10.1214/aos/1176342360
发表时间: 1973-01-01
影响因子: 4.5
作者:
FERGUSON, TS
通讯作者: FERGUSON, TS