Hedibert Freitas Lopes conducts research in Markov Chain Monte Carlo techniques and Sequential Monte Carlo methods applied to multivariate econometrics and time-series models; modeling time-varying covariance of multivariate time series through latent factor analysis; Choleski decomposition and other factorizations; dynamic models and Bayesian inference; and computation. He is mainly interested in the implementation of the Bayesian paradigm to solve real large-scale problems in Econometrics and other fields of Economics.
"My research highlights the importance of model uncertainty and how it can be measured and accounted for based on modern computational statistics schemes," he said. Students should take away from his classes a broader, deeper, and more responsible understanding of the vast applicability of statistical methods and thinking when dealing with complex real problems involving several sources of uncertainty.
His research has been published in a number of academic journals and books and has won him fellowships from Coordencao de Aperfeicoamento de Pessoal de Nivel Superior, the Research Institute for Applied Economics, and Conselho Nacional de Desenvolvimento Cientifico e Tecnologico.
Lopes serves on several masters and PhD committees and referees for journals including Journal of Econometrics, Journal of Applied Econometrics, Journal of Financial Econometrics, Econometrics Journal, Journal of the American Statistical Association, Journal of the Royal Statistical Society, Journal of Computational and Graphical Statistics, Biometrics, and Quantitative Marketing and Economics.
Lopes earned his bachelor's degree in Statistics in 1991 and master's degree in Statistics in 1994 from Federal University of Rio de Janeiro. He also earned a master's degree in Statistics and Decision Sciences in 1998 and a PhD in 2000 from Duke University.
Outside of the classroom, Lopes enjoys spending time with his family and friends, reading, jogging, and dancing.
With Carvalho, C., Johannes, M. and Polson, N., "Particle learning and smoothing," Statistical Science (2010).
With Polson, N., "Extracting SP500 and NASDAQ volatility: The credit crisis of 2007-2008," Handbook of Applied Bayesian Analysis (2010).
With Salazar, E. and Gamerman, D., "Spatial dynamic factor models," Bayesian Analysis, 3, 759-792 (2008).
With Gamerman, D., Markov Chain Monte Carlo: Stochastic Simulation for Bayesian Inference (2nd ed.), Chapman & Hall/CRC (2006).
With West, M., "Bayesian Model Assessment in Factor Analysis," Statistica Sinica, 14, 41-67 (2004).
For a listing of research publications please visit Hedibert Freitas Lopes’s university library listing page
REVISION: Particle Learning and Smoothing
Date Posted: Nov 09, 2011
Particle learning (PL) provides state filtering, sequential parameter learning and smoothing in a general class of state space models. Our approach extends existing particle methods by incorporating the estimation of static parameters via a fully-adapted filter that utilizes conditional sufficient statistics for parameters and/or states as particles. State smoothing in the presence of parameter uncertainty is also solved as a by-product of PL. In a number of examples, we show that PL outperforms
REVISION: Tracking Flu Epidemics Using Google Flu Trends and Particle Learning
Date Posted: Mar 17, 2010
In the second half of 2009 the world experienced an intense influenza activity. The new 2009 H1N1 virus, formerly known as the swine flu, has in only five months found its way from Mexico to a majority of the countries on the planet. The fears of a large second-wave pandemic and its potential impact on health and economic outcomes have underlined the importance of accurate and fast disease surveillance mechanisms capable of suggesting timely public health interventions.
In this paper we introdu
Bayesian Model Uncertainty In Smooth Transition Autoregressions
Date Posted: Feb 21, 2006
In this paper, we propose a fully Bayesian approach to the special class of nonlinear time-series models called the logistic smooth transition autoregressive (LSTAR) model. Initially, a Gibbs sampler is proposed for the LSTAR where the lag length, k, is kept fixed. Then, uncertainty about k is taken into account and a novel reversible jump Markov Chain Monte Carlo (RJMCMC) algorithm is proposed. We compared our RJMCMC algorithm with well-known information criteria, such as the Akaike information