Jim LeSage, Professor of Economics, Publishes Econometrics Paper with Undergraduate Student

An article by Dr. James P. LeSage, Fields Endowed Chair in Urban and Regional Economics, and Mr. Daniel Hendrikz entitled: “Large Bayesian Vector Autoregressive Forecasting for Regions: A Comparison of Methods Based on Alternative Disturbance Structures,” was recently accepted for publication in the Annals of Regional Science. Mr. Hendrikz was a McCoy College undergraduate student who completed his degree in December 2018. The research resulted from work done by Dr. LeSage and Mr. Hendrikz during Summer 2018 and Fall 2018. Mr. Hendrikz was a student in Econometrics in Spring 2018, and the research extends methods used in a classroom forecasting competition. Dr. LeSage and Mr. Hendrikz found that extending conventional Bayesian vector autoregressive forecasting models to allow for non-Gaussian, heteroscedastic and time-dependent disturbance structures resulted in improved forecast accuracy and forecast precision of state-level economic activity in 46 of 48 states. Experimental forecasts were carried out for the lower 48 US states over a six-year period following the recession of 2008. The article also explored the role of spatial prior information regarding economic activity in neighboring states as well as non-spatial prior information pertaining to past period state-level economic activity. Dr. LeSage has published six articles with Texas State students over the past 10 years that have received over 300 citations. (View the citations for all Dr. LeSage’s publications here.)