An asymptotic analysis of generative, discriminative, and pseudolikelihood estimators

Statistical and computational concerns have motivated parameter estimators based on various forms of likelihood, e.g., joint, conditional, and pseudolikelihood. In this paper, we present a unified framework for studying these estimators, which allows us to compare their relative (statistical) efficiencies. Our asymptotic analysis suggests that modeling more of the data tends to reduce variance, but at the cost of being more sensitive to model misspecification. We present experiments validating our analysis.

• Type: Conference
• Year: 2008
• Where: ICML
• Authors: Percy Liang, Michael I. Jordan
• Keywords: Asymptotic Analysis, ICML 2008, Machine Learning, Parameter Estimators