General Interests: Econometrics, Statistical Learning, Behavioral Economics, Ubiquitous Computing, Agricultural Economics.
My research roadmap provides a detailed guide to my research plans.
My research interests are centered around a fundamental issue that I see in the economics field: computer scientists are increasingly able to predict economic trends better than economic models do. That economic models often do not correspond to reality is so well known that there are many jokes about it. While abstract economic models have provided insights into the fundamental interactions of people, good, and markets, they have also consistently failed to predict human behavior on both the micro and macro scale.
This is a problem because we want to not only be able to predict well, but also to understand the underlying relationships of an economic system. Behavioral economics has made headway in this area, but new models of eonomics have failed to materialize. In truth, the complex systems of economics may be difficult and/or impossible to reliably predict. But this should not stop us from making some headway.
The use of large datasets and statistical learning techniques to help answer causal questions in economics can help to revise models to provide a better understanding of how humans and the economic systems they create behave. Some work by Hal Varian (2014) and Athey and Imbens (2015) is happening in this area, and I anticipate much more to follow.
- Varian, Hal R. "Big data: New tricks for econometrics." The Journal of Economic Perspectives 28, no. 2 (2014): 3-27.
- Athey, Susan, and Guido Imbens. "Machine learning methods for estimating heterogeneous causal effects." arXiv preprint arXiv:1504.01132 (2015).