Statistics Seminar（2018-23）

Topic: Integrated Deviance Information Criterion for Latent Variable Models

Speaker: Yong Li(李勇), Renmin univesrity of China

Time: Thursday, December 20, 14:00-15:00

Place: Room K02, Guanghua Building 2

Abstract:

Deviance information criterion (DIC) has been widely used for Bayesian model comparison, especially after Markov chain Monte Carlo (MCMC) is used to estimate candidate models. This paper studies the problem of using DIC to compare latent variable models after the models are estimated by MCMC together with the data augmentation technique. Our contributions are twofold. First, we show that when MCMC is used with data augmentation, it undermines theoretical underpinnings of DIC. As a result, by treating latent variables as parameters, the widely used way of constructing DIC based on the conditional likelihood, although facilitating computation, should not be used. Second, we propose two versions of integrated DIC (IDIC) to compare latent variable models without treating latent variables as parameters. The large sample properties of IDIC are studied and an asymptotic justification of IDIC is provided. Some popular algorithms, such as the EM, Kalman and particle filtering algorithms, are introduced to compute IDIC for latent variable models. IDIC is illustrated using asset pricing models, dynamic factor models, and stochastic volatility models.

Introduction:

Dr Yong Li is Professor of Statistics and Finance in Renmin univesrity of China. His research interest is Bayesian Financial Econometrics. He has published more than 40 journals in international and Chinese journals such as Journal of Econometrics, Economic research or management word.  He is  the Young Changjiang Scholar of Education Administration (Econometrics).  

Your participation is warmly welcomed!