Call for Proposals: Big Data and High-Performance Computing for Financial Economics Training Workshop

Big data, novel statistical tools, and new model-solving techniques provide opportunities for financial economists to explore new topics and to answer traditional questions in novel and move revealing ways. Unlocking the power of these new tools, however, is often beyond the reach of standard computing resources. High-performance computing resources can not only help economists pursue existing research topics more efficiently, but can also foster entirely novel lines of research. The support the emerging research community that is harnessing the power of high-performance computing in financial economics, the NBER will host a one-day workshop that will combine lectures on tools and methods with a discussion of early-stage research projects.

This workshop and research seminar, which will be organizes by Toni Whited of the University of Michigan and NBER, and Mao Ye of the University of Illinois and NBER, will be held in Cambridge, MA on Saturday, July 14, 2018. This seminar will be supported by the Extreme Science and Engineering Discovery Environment (XSEDE) program of the National Science Foundation.

The meeting will have two parts. The morning will be devoted to lectures on tools, methods, and strategy for the use of high-performance computing, while the afternoon will be focused on discussion of research projects. There will be three presentations in the morning:

Computation Intensive Research Opportunities: Whited will describe a range of topics that are made possible by XSEDE resources, such as such as estimating and testing dynamic models in corporate finance using simulated method of moments, indirect inference, and the use of model empirical policy functions.

Data Intensive Research Opportunities: Ye will describe research opportunities that exploit large data sets, such as the study of high-frequency trading, automated trading, the use of machine learning and social media in asset pricing, and the development of theoretical models inspired by the analysis of big data.

Computational and Data-Analytic Tools: Experts from XSEDE will describe XSEDE resources and the basics of high-performance computing and big data techniques for social sciences. They will assist conference participants with their application for a startup or research allocation, as well as their applications for Extended Collaborative Support Services (ECSS) funded by XSEDE.

The NBER invites submissions of research proposals that could be discussed as part of the afternoon session. Projects must leverage high-performance computing resources, large data storage capacity, and/or new large data sets. The goal of the meeting is to provide researchers who are exploring questions that would benefit from high performance computing with an opportunity to learn about resources in this field, and to interact with other researchers who are interested in similar questions.

Potential topics for presentation are discussion include, but are not limited to:

* Data of large size, such as trade and quote market microstructure data.

* Unstructured data, such as data from social media, webpages and images.

* Applications of machine learning and/or artificial intelligence in finance.

* Estimating models with high-dimensional fixed effects.

* Calibrating and estimating structural models in financial economics.

* Solving high-dimensional dynamic programming problems.

Researchers who are working on problems such as these, for which high-performance computing could enhance the research project, are encouraged to submit a two-page proposal at the following website:

Proposals must be submitted by midnight EDT on Wednesday April 25 2018. Proposals from researchers with and without NBER affiliations are welcome, as are proposals from early career scholars and from researchers who are members of groups that are historically under-represented in economics and finance.

The proposal should include an abstract describing the research objectives, a discussion of the empirical or theoretical research strategy and the computing language to be used for the analysis, for empirical work, a description of the dataset and its size, and an explanation of why the project requires XSEDE resources. For example, if the project requires parallel computing to speed up the data analysis, or requires large amounts of memory for estimation or simulation, that should be explained. In addition, researchers who are interested in getting computing resources prior to the conference can dual-submit their proposal to the XSEDE startup allocation website. The organizers will select up to eight proposals for presentation and discussion at the meeting.

Decisions about which proposals are accepted will be announced in early May. NBER will cover the hotel and travel cost for two authors per selected proposal. Other participants may also attend the meeting if space permits, but priority will be given to those who submit proposals. Anyone interested in attending the meeting, but who does not have a proposal, should email Carl Beck in the NBER Conference Department

For questions about the subject matter of the meeting, please contact Mao Ye For questions about conference logistics, please contact Carl Beck in the NBER's Conference Department

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us