Theoretical Properties of Composite Likelihoods

• 作者: Xiaogang Wang Yuehua Wu
• 来源期刊: 统计学期刊(英文)
• 学科: 数学
• 关键词: Composite LIKELIHOOD I-Divergence Information Theory LIKELIHOOD WEIGHTS MAXIMUM ENTROPY Distribution
• 年，卷（期）: tjxqkyw_2014,（3）
• 页码范围: 188-197
• 页数: 10页
• 分类号: O1

摘要

The general functional form of composite likelihoods is derived by minimizing the Kullback-Leibler distance under structural constraints associated with low dimensional densities. Connections with the I-projection and the maximum entropy distributions are shown. Asymptotic properties of composite likelihood inference under the proposed information-theoretical framework are established.