Akaike's information criterion and recent developments in information complexity
Akaike's information criterion and recent developments in information complexity
复制标题
DOI：
10.1006/jmps.1999.1277
复制
发表时间：
2000-03-01
影响因子：
1.8
通讯作者：
Bozdogan, H
中科院分区：
文献类型：
--
作者：
Bozdogan, H
本文简要研究了Akaike（1973）的信息准则（AIC）的基本思想。然后，我们介绍了 Bozdogan（1988a、1988b、1990、1994d、1996、1998a、1998b）模型选择的新熵或信息复杂性（ICOMP）标准的一些最新进展。 ii ICOMP 作为模型选择标准的基本原理是，它将拟合不良项（例如减去最大对数似然的两倍）与模型的复杂性度量相结合，这与 AIC 或其变体不同。通过考虑参数的相互依赖性，估计 us ut ll 作为模型残差的依赖性。我们根据估计的逆费舍尔信息矩阵对整体模型复杂性概念的量化，对 ICOMP 的一般形式进行了操作。该方法产生两个 Kullback-Leibler 距离之和的近似值。使用复杂性的相关形式，我们进一步提供了另一种形式的 ICOMP 来考虑模型参数估计之间的相互依赖性（即 Lon 相关性）。随后，我们通过提供几个真实的以及 Monte Carte 模拟的例子来说明这种新模型选择标准的实用性和重要性：并将 itu 性能与 AIC 进行比较。或其变体，(C) 2000 学术出版社。
In this paper we briefly study the basic idea of Akaike's (1973) information criterion (AIC). Then, we present some recent development on a new entropic or information complexity (ICOMP) criterion of Bozdogan (1988a, 1988b, 1990, 1994d, 1996, 1998a, 1998b) for model selection. ii rationale for ICOMP as a model selection criterion is that it combines a badness-of-fit term (such as minus twice the maximum log likelihood) with a measure of complexity of a model differently than AIC, or its variants. by taking into account the interdependencies of the parameter estimates us ut ll as the dependencies of the model residuals. We operationalize the general form of ICOMP based on the quantification of the concept of overall model complexity in terms of the estimated inverse-Fisher information matrix. This approach results in an approximation to the sum of two Kullback-Leibler distances. Using the correlational form of the complexity, we further provide yet another form of ICOMP to take into account the interdependencies (i.e.. Lon correlations), among the parameter estimates of the model. Later, we illustrate the practical utility and the importance of this new model selection criterion by providing several real as well as Monte Carte simulation examples: and compare itu performance against AIC. or its variants, (C) 2000 Academic Press.