期刊
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
卷 99, 期 465, 页码 216-227出版社
AMER STATISTICAL ASSOC
DOI: 10.1198/016214504000000214
关键词
cross-validation; generalized linear model; goodness of fit; jackknife; lack of fit; likelihood ratio test; parametric bootstrap; sandwich matrix
A new test of model misspecification is proposed, based on the ratio of in-sample and out-of-sample likelihoods. The test is broadly applicable and, in simple problems, approximates well-known, intuitive methods. Using jackknife influence curve approximations, it is shown that the test statistic can be viewed asymptotically as a multiplicative contrast between two estimates of the information matrix, both of which are consistent under correct model specification. This approximation is used to show that the statistic is asymptotically normally distributed, although it is suggested that p values be computed using the parametric bootstrap. The resulting methodology is demonstrated with various examples and simulations involving both discrete and continuous data.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据