Econometric Game 2011
On the 12th, 13th and 14th of April 2011, Amsterdam was the décor for the most prestigious competition in econometrics amongst students; the Econometric Game 2011. Sponsored by ING, Nationale Nederlanden and supported by the University of Amsterdam, the student association for actuarial sciences, econometrics and operational research (VSAE) of the University of Amsterdam organized the 12th Econometric Game. Because of the growing popularity of the Econometric Game among students, professors and business people all over the world a strong field was composed of prestigious universities as the University of Oxford and University of Cambridge, five Dutch universities and new participants like the Indian Institute of Technology Kanpur. Professor Windmeijer (University of Bristol) was prepared to make an interesting case: the effects of alcohol usage during pregnancy on the performance of children in school. Genetic predisposition was an important instrument in the investigation.
The Econometric Game started off on a beautiful Tuesday in the Duif, a former church in the city centre of Amsterdam. Over 200 students in econometrics gathered to attend the official opening of the Econometic Game 2011. Presentations of Gisella van Vollenhoven (ING) and Dymph van den Boom (rector magnifica UvA) introduced the subject after which the Professor Windmeijer announced the case. After the introduction, 125 talented students in 25 teams from all over the world settled down to start solving the case. The Duif, transformed into blue and white, was the workplace during the first two days. Wearing their Econometric-Game-sweaters the participants were spread over the tables with their books, notes, laptops, cables, calculators, energy-drinks and more so they could achieve the best result. During the first day, the participants had the opportunity to orientate themselves on the case with articles which were provided by the committee and material they found on their own. At the end of the day the teams could relax with some drinks, a dinner and they were able to socialize with each other and with employees of ING and NN.
After a day of introduction and orientation on the subject the competition really started on the second day. The teams received the datasets to start their real analyses. They had to find a solution and come up with a report before the end of the day. Despite the enormous time pressure, the committee kept to the deadline: at exactly 4.00PM the reports had to be handed in. After that it was waiting for the results. At 23:00 the expert jury, containing renowned professors, had checked all the rapports and decided which ten teams were good enough to continue to day three of the Econometric Game. Frank Windmeijer announced the finalists and the room was a mix of cheering and disappointment. The upside for the eliminated teams was that they could enjoy another drink in the bar of the Hotel Arena.
On the final day the finalists were working on a second case in the Academical Medical Centre (AMC). The other participants could enjoy themselves on a canal cruise through Amsterdam. After some sightseeing, the Econometric Game Congress 2011 took place. More than 200 attendees, including the eliminated teams and interested students from the University of Amsterdam and the Vrije University of Amsterdam, watched different professors speak about their specialism. Meanwhile, the finalists finished the case in the AMC, after which they presented their findings for the second case at the congress. Around midnight the moment of truth for the finalist came: the announcement of the winning team. The jury had decided which team was best in solving the case and presenting it. Top executive Ricardo Sookdeo (ING) and chairman of the jury Frank Windmeijer had the honor to present the winner: Maastricht University. The room applauded while they opened the champagne bottle and received the Cup. The partying started and went on till the early hours. It was a successful ending to a successful event.
The assignment of the Econometric Game 2011 was to identify and estimate the causal effect of maternal alcohol consumption during pregnancy on cognitive outcomes of the child, using a genetic marker as instrument.
Mendelian Randomisation is the term used to describe the random allocation of genes from parents to offspring. Recent advances in genotyping have made it possible to obtain genetic information more easily and economists are increasingly interested in the use of genetic variants as instrumental variables to identify the effect of a modifiable non-genetic risk factor on the outcomes of interest. Of course, there may be various reasons why genetic marker may not be a valid instrument. Also, a genetic marker may only be weakly associated with the risk factor.
In the case of the Econometric Game 2011 we contributed to the long and ongoing debate about whether alcohol consumption by the mother during pregnancy has a beneficial or detrimental effect on the development of the foetus. The data set that was analysed contained information about the mother’s alcohol intake during pregnancy and the children’s school test scores at age of 14. A simple regression analysis may not identify the causal effect of alcohol intake due to unobserved factors that affect both maternal alcohol intake and the human capital formation of the child. We had information about the mother on the presence of a genetic marker which has been shown tob be associated with alcohol consumption and may therefore be a good candidate for use as an instrumental variable. This genetic marker may thus enable us to identify/estimate a causal effect of maternal alcohol consumption on the child’s cognitive ability by means of instrumental variables estimation techniques.
Casemaker and jury
The casemaker in 2011 was prof. F. Windmeijer.
Frank Windmeijer is a professor of Econometrics at the University of Bristol since July 2005. After the completion of his PhD on goodness-of-fit measures at the University of Amsterdam in 1992, he first became a visiting lecturer at the Australian National University, before moving to London in 1994. There he first held a Marie Curie fellowship at University College London and then became a senior research fellow at the Institute for Fiscal Studies and a co-director of the Centre for Microdata Methods and Practice. His best known econometrics contribution is the finite sample correction to the two-step GMM standard errors, as programmed up in e.g. Stata. He has further contributed to various applied projects, including analysing demand for health care. In recent years he has started to collaborate with colleagues in the Social Medicine department, assessing the usefulness of genetic markers as instruments for e.g. obesity or alcohol intake.
The jury consists of dr. C.G.H. Diks, dr. N.P.A. van Giersbergen, dr. S. von Hinke Kessler Scholder and prof. D.A. Jaeger. The case maker prof. F. Windmeijer is also part of the jury.
Dr. Cees Diks says that the nice thing about having a background in time series analysis is that it can be applied in virtually any field. He has worked on various applications, including heart data (ECGs), brain data (EEGs) and hydrological data. He knows from experience that working on these data is peanuts compared to the challenges econometricians face when modelling economic or financial econometric data. The task to find structure in a ‘non-stationary sea of noise’ is often huge.
After completing his PhD at the UvA in the area of the bootstrap, dr. Noud van Giersbergen worked for several years outside of academia. First, he was a researcher at Statistics Netherlands, but after a year he joined Watson Wyatt as an actuarial consultant. Finally, he returned to the UvA to teach and do research. Dr. Van Giersbergen says that he finds teaching very rewarding. Discovering new results in theoretical research can be very exciting, but over the years his research also includes empirical work. Although the results obtained by empirical analysis are less general than in theoretical theory, the conclusions can sometimes be of vital importance in practice.
Dr. Stephanie von Hinke Kessler Scholder is a post-doctoral research fellow at the Healthcare Management Group, Imperial College London. Her research draws on a range of applied microeconometric methods relating to questions within health economics, with a particular focus on the economics of (childhood) obesity. She is interested in the use of genetics within economics research; her ongoing collaboration with economists and genetic epidemiologists uses specific genetic variants as instrumental variables for fat mass, height, alcohol consumption and smoking behaviour to examine their effects on different child and adult economic outcomes.
David A. Jaeger received his BA from Williams college (1986), and his MA (statistics, 1993) and PhD (economics, 1995) from the University of Michigan. In 1995, he was the first recipient of the W. E. Upjohn Institute Dissertation Award, and in 2003/2004 he received a fellowship from the Alexander von Humboldt Stiftung. His research focuses on immigration and migration, education, conflict and applied econometrics. His work has been published in the American Economic review, the Journal of the American Statistical Assocation, the Review of Economics and Statistics, the Journal of Labor Economics, and the Quarterly Journal of Political Science, among others.
The top 3 of the Econometric Game 2011 were:
- Maastricht University
- University of Oxford
- University of Cambridge
We are working are publishing their findings.
From left to right: Ewout Schotanus, Mariëlle Coenen, Julien Reincke, Ilona van Mechelen, Jeroen Does, Myrna Hennequin and Rick van der Peet.