New Professor Federico Bugni

New Professor Federico Bugni

08 June 2010 12:00AM

It’s not uncommon to hear economics students complaining about their struggles with econometrics, but Federico Bugni found himself attracted to the subject as a graduate student at Northwestern University.

“I enjoy technical work,” said the department’s new professor and econometrician. “Econometrics is extremely useful to provide answers to all sorts of economic problems.”

Econometrics is the application of statistical analysis to the study of economics.

“For example, if there were a situation in which a firm had a monopoly, and a new firm then entered the market, economics would tell you that profits at the first firm should decline,” Bugni explained. “Now in order to quantify the amount of this decline, you need to rely on econometrics.”

The use of statistical tools to estimate the parameters of economic models allows us to provide quantitative answers to our problems. Currently, Bugni is focusing on a type of econometric analysis called partial-identification. A partially identified model is a special kind of econometric model used to analyze data in the presence of missing observations or multiplicity of equilibria.

“For example, in labor economics, surveys are often conducted – and practically every survey has some missing data. How are you going to handle that issue?” asks Bugni. “In the past, economists have had to make what some may consider unreasonable assumptions, but now partial-identification can help address this issue.”

To illustrate, in a project with colleague Joe Hotz and Ph.D. student Esteban Aucejo, Bugni examines the determinants of college graduation. Using data from student surveys from the University of California system, the researchers use econometrics to quantify the effect of income and other characteristics on the probability of graduation.

“During the time period we are looking at, the universities in California were affected by new state legislation which mandated that race cannot be used as a factor in admissions,” explained Bugni. “The number of students who declined to report race in surveys increased tremendously, so we are missing nearly 20% of our data on race. It would be unreasonable to assume that those students who refused to answer the race question are of the same race as those individuals who did answer the question. This is a perfect example of the kind of issue that a partially identified econometric model can help resolve.”

In addition to partial-identification, Bugni’s research interests include theoretical econometrics, inference, moment inequalities, missing data and stochastic processes. He said he’s happy to have a senior colleague, Shakeeb Khan, who does similar econometrics work.

Bugni also expresses a desire to team up with faculty in other specialties. “I enjoy collaborating with the faculty working in Applied Microeconomics. They are working on problems in industrial organization and labor economics which I find very exciting,” Bungi said.

Before joining the faculty at Duke, Bugni earned his Ph.D. in economics from Northwestern University. Previously, he received Master’s and bachelor’s degrees in economics from the Universidad de San Andres in Buenos Aires, Argentina, his hometown.

Bugni’s work is already gaining attention in his field. He has published two articles, “Goodness-of-fit tests for functional data” in The Econometrics Journal and “Bootstrap inference in Partially Identified Models Defined by Moment Inequalities: Coverage of the Identified Set” in Econometrica, widely regarded as a leading economics journal.

“My work is interesting and challenging,” Bugni said with a smile. “I hope to contribute to the department and my field.”

Learn more about Professor Bugni at his profile page.