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.
关键词：Composite Likelihood; I-Divergence; Information Theory; Likelihood Weights; Maximum Entropy Distribution
