Issue link: https://www.e-digitaleditions.com/i/54913
research How much can business students learn from behav- ioral economics? One of the big lessons from behavioral economics and social science is how little we know. If we take this les- son to heart, most of our students should learn to be humble and realize that the answer to most questions should be, "I don't know, but I have the tools to test it." But imagine if a student came to study with me for two years, and at the end of those two years, the only thing he had learned was that he didn't know much, but he had the tools! It would be a more honest educa- tion, but incredibly unsatisfying for the student. What are the most important tools they should have before graduation? I think we need to teach students more statistics. They need to know how to use existing data and how to generate new data in case the existing data does not provide an answer. The world works in statistical probabilities. As researchers, we generate a lot of data, about how compensation works, about what really motivates people, and who gets motivated by what. But we don't transmit that information nearly enough. What project are you working on now? What's next? When I started writing, I wanted to write a cookbook called Dining Without Crumbs: The Art of Eating over the Sink. In it, I would talk about how to think about life through the lens of the kitchen, the place where we cre- ate and destroy and think about others and procrastinate. But nobody wanted to publish this. They said, "Why don't you write a book about your research?" That was how Predictably Irrational was born. Then I wanted to write my cookbook, and they said, "How about one more on your research?" So I wrote The Upside of Irra- tionality. But now I'm ready to write my cookbook! How would you like schools to use your work? Well, there was one professor who took my advice a bit too seriously. He had read my research that showed that when people were asked to recall the Ten Com- mandments, they cheated less. So, he had his students read the following: "If I cheat on this exam, I will regret it for the rest of my life, and I will lose my eter- nal soul." He got into lots of trouble. So, you want schools to use your work rationally? That's right. 62 May/June 2011 BizEd Weatherproofing the Financial Markets THE SAME METHODS meteorologists use to predict the weather might also be used to make more accu- rate predictions of the stock market, according to Ash- ish Tiwari, associate professor of finance; Gene Savin, professor of economics; and doctoral student Michael O'Doherty at the University of Iowa's Tippie College of Business in Iowa City. Building on the work of a former Tip- pie economics colleague, John Geweke, they studied how meteorologists incorporate dif- ferent types of climate information and then synthesize the results of each to determine whether a region is more likely to expect rain or sunshine. Tiwari, Savin, and O'Doherty recently tested how well a series of model pools pre- dicted stock returns between 1932 and 2008. They used the single-factor capital asset pricing model, the Fama-French three-factor model, and the Carhart four-factor model— the "workhorse models" of empirical asset pricing, according to Tiwari. After comparing the prediction to reality, they found that using two models led to more accurate predictions than using a single one. Moreover, using all three models was more accurate still. Eventually, they added two more models to create a five-model pool. Because each one compensates for weaknesses in the oth- ers, the pool captures more information than any single approach, they explain. The only requirement is that each new model contain unique information. Michael O'Doherty Ashish Tiwari Multiple model pools could more accu- rately predict not only the markets, but also economic systems such as the real estate mar- ket, unemployment rates, or GDP. Currently, the three are looking at how model pools can improve the benchmarks used to evaluate the performance of mutual fund and hedge fund managers. The paper "Modeling the Cross Section of Stock Returns: A Model Pooling Approach" can be down- loaded from SSRN at ssrn.com/abstract=1570772. Gene Savin