Areas of Expertise (13)
Carlos M. Carvalho is a statistician, researcher, and educator who uses Bayesian statistics to address high-dimensional problems from a variety of fields, including finance and capital markets, genetics, health care, and political campaigns.
Carvalho is an associate professor of statistics in the department of Information, Risk, and Operations Management at the McCombs School of Business, The University of Texas at Austin.
Carvalho was previously on the faculty at The University of Chicago Booth School of Business, Duke University, and IBMEC in Rio de Janeiro, Brazil.
He is a CBA Foundation Advisory Council Centennial Fellow, since 2012, and he was awarded The Donald D. Harrington Fellowship by UT Austin in 2009. He has been an analytics consultant with Dell, Inc. since 2013.
His current research includes:
1) Advanced statistics and econometrics in asset pricing problems;
2) Causal inference in high dinemsional settings;
3) Dimensionality reduction in large-scale multivariate problems;
4) Sparse models for high-dimensional covariance matrices;
5) Graphical models and sparse factor models;
6) Model search/selection in linear models and graphical models;
7) Dynamic graphical models in multivariate financial time series and portfolio analysis;
8) Conditional variance models and multivariate stochastic volatility;
9) Sequential estimation and particle filtering; and
10 Parallel statistical computation.
Duke University: Ph.D., Statistics 2006
Thesis: "Structure and Sparsity in High-Dimensional Multivariate Analysis".
Federal University of Rio de Janeiro: M.Sc., Statistics 2002
Thesis: "Bayesian Analysis of Stochastic Volatility Models with Multiple Regimes".
IBMEC Business School: B.Sc., Economics 1999
Rio de Janiero, Brazil
Media Appearances (1)
The Science Behind Big Data
OPEN Magazine print
If there’s a statistician’s version of a Swiss Army knife, it might be Bayesian analysis. A technique for finding patterns in complex systems, Carvalho first used it to pinpoint genes that affect a cancer patient’s chances of recovery.
Listing of top scholarly works by Carlos M. Carvalho.
This article revisits the venerable problem of variable selection in linear models.
In this paper, we present two dynamic cardiac risk estimation models, focusing on different temporal signatures in a patient’s risk trajectory.
We study the purchasing power parity (PPP) puzzle in a multi-sector, two-country, sticky-price model.
We show that Particle Learning (PL) outperforms existing particle filtering alternatives and proves to be a competitor to MCMC.
This paper introduces sectoral heterogeneity in price stickiness into an otherwise standard sticky price model to study how it affects the dynamics of monetary economies.