CandidatesBack to Candidates
Job Market Paper
This paper constructs unit-specific density forecasts for a panel of firms or households using a dynamic linear model with common and heterogeneous parameters and cross-sectional heteroskedasticity. The distribution of the heterogeneous coefficients is modeled nonparametrically, allowing for correlation between heterogeneous coefficients and initial conditions as well as unit-specific regressors. A benchmark for the density forecasts is the (infeasible) oracle forecast which is defined as the posterior predictive distribution for the unit-specific outcomes under the assumption that the common parameters and the distribution of the heterogeneous coefficients are known. I develop a simulation-based posterior sampling algorithm specifically addressing the nonparametric density estimation of unobserved heterogeneous coefficients. I prove that both the estimated common parameters and the estimated distribution of the heterogeneous coefficients achieve posterior consistency, and that the density forecasts asymptotically converge to the oracle forecast. Monte Carlo simulations demonstrate improvements in density forecasts relative to parametric approaches. An application to young firm dynamics also shows that the proposed predictor provides more accurate density predictions.
joint with Hyungsik Roger Moon and Frank Schorfheide
We consider the problem of forecasting panel data with a large cross-sectional and a small time-series dimension. We consider a linear correlated random effects specification and construct a predictor using Tweedie's formula for the posterior mean of the heterogeneous coefficients. This formula utilizes cross-sectional information to transform the unit-specific (quasi) maximum likelihood estimator into an approximation of the posterior mean under a prior distribution that equals the population distribution of the random coefficients. We show that the risk of a predictor based on a non-parametric estimate of the Tweedie correction is asymptotically equivalent to the risk of a predictor that treats the correlated-random-effects distribution as known (ratio-optimality). Our empirical Bayes predictor performs well compared to various competitors in a Monte Carlo study. In an empirical application we use the predictor to forecast revenues for a large panel of bank holding companies and compare forecasts that condition on actual and severely adverse macroeconomic conditions.
joint with Mert Demirer, Francis X. Diebold, and Kamil Yilmaz
Revise and Resubmit, Journal of Applied Econometrics
We use Lasso methods to shrink, select and estimate the network linking the publicly-traded subset of the world's top 150 banks, 2003-2014. We characterize static network connectedness using full-sample estimation and dynamic network connectedness using rolling-window estimation. Statically, we find that global banking connectedness is clearly linked to bank location, not bank assets. Dynamically, we find that global banking connectedness displays both secular and cyclical variation. The secular variation corresponds to gradual increases/decreases during episodes of gradual increases/decreases in global market integration. The cyclical variation corresponds to sharp increases during crises, involving mostly cross-country, as opposed to within-country, bank linkages.
Financial institution networks potentially feature large structural changes over time, which would affect the systemic risk of the whole system. This paper focuses on the Diebold-Yilmaz connectedness measure obtained via variance decomposition, and provides a fused Lasso method to estimate structural changes in the VAR coefficients. To address the high-dimensionality problem along both cross-sectional and time-series dimensions, the fused Lasso estimator penalizes the VAR coefficients as well as their successive differences. I prove that under reasonably general conditions, the proposed method can consistently detect the unknown number of breaks, the estimated break dates are sufficiently close to the true dates, and the estimated coefficients asymptotically converge to the true values. Monte Carlo simulation evidence is presented, along with an application to stock return volatilities of the major financial institutions traded in the U.S. stock market. Results show that structural changes in the interaction pattern are more responsible for the recent financial crisis, while the effects of unfavorable individual shocks are negligible.
joint with Francis X. Diebold and Kamil Yilmaz
We use variance decompositions from high-dimensional vector autoregressions to characterize connectedness in 19 key commodity return volatilities, 2011-2016. We study both static (full-sample) and dynamic (rolling-sample) connectedness. We summarize and visualize the results using tools from network analysis. The full-sample results reveal clear clustering of commodities into groups that match traditional industry groupings, but with some notable differences. The energy sector is most important in terms of sending shocks to others, and energy, industrial metals, and precious metals are themselves tightly connected.
PIER Workshop on Quantitative Tools for Macroeconomic Policy Analysis, Teaching Assistant (May 2016)
Topics in Econometrics: Forecasting, Teaching Assistant (Fall 2014)
Econometrics II: Methods (graduate), Teaching Assistant (Spring 2013, 2014)
Introduction to Microeconomics, Recitation Instructor (Fall 2013)
Intermediate Microeconomics, Recitation Instructor (Fall 2012)
Macroeconomics (graduate), Teaching Assistant (Fall 2010)
Federal Reserve Bank of Philadelphia, Philadelphia, PA (December 2016)
Midwest Econometrics Group Annual Meeting, Champaign, IL (October 2016)
Federal Deposit Insurance Corporation (FDIC), Washington, DC (November 2015)
EconCon Conference, Princeton, NJ (as presenter and discussant) (August 2014)
Federal Reserve Bank of Richmond, Richmond, VA (June, August 2014)
International Journal of Central Banking
Honors, Scholarships, and Fellowships:
Robert Summers Dissertation Fellowship in Economics, University of Pennsylvania (2016)
Maloof Family Dissertation Fellowship in Economics, University of Pennsylvania (2015)
Xinmei Zhang and Yongge Dai Fellowship, University of Pennsylvania (2012)
Best Performance in Econometrics 1st Year, University of Pennsylvania (2012)
University Fellowship, University of Pennsylvania (2011)
Norman M. Kaplan Memorial Prize (best first-two-year performance), University of Rochester (2010)
Lionel and Blanche McKenzie Family Fellowship (best 1st-year performance), University of Rochester (2009)
Francis X. Diebold
Francis J. DiTraglia
I am on the job market and will be available for interviews during the AEA meetings in Chicago on January 6 to January 8, 2017.