Workshop on Machine Learning in Finance to Feature Leading Thinkers in the Field
The Data Science Institute (DSI), Bloomberg and Columbia Engineering are hosting the 5th annual workshop on Machine Learning in Finance, a daylong event featuring leading thinkers in the field.
The workshop, scheduled for May 17 at Lerner Hall, is sponsored by the Financial and Business Analytics Center at DSI, and Columbia’s Center for Financial Engineering. Boxed lunches will be provided; there will also be coffee breaks throughout the day and a wine and cheese reception will follow the event. Online registration is available HERE.
The day will be divided into eight workshop sessions, one of which will be given by Simona Abis, an assistant professor at the Columbia Business School and a DSI member. Her talk, "The Informational Content of Mutual Fund Prospectuses," will discuss the drafting of prospectuses for mutual funds and why she believes the Securities Exchange Commission (SEC) has not demonstrated their relevance for investors.
Abis uses machine learning to study the importance of the textual information contained in prospectuses of a broad sample of equity mutual funds, which are available through EDGAR, the SEC’s online reporting system. Using supervised learning on ex-ante (forecast in advance) prospectuses, she can predict which funds are more likely to engage in destructive, agency-related risk-shifting behavior. She also uses unsupervised learning to group funds into distinct clusters based on the similarity in prospectus text, uncovering groups with natural interpretations such as “quantitative,” “macro-focused,” and “derivatives risk.” Abis has found that membership in particular clusters is predictive of the shape of funds’ return distributions.
The other morning workshops include Kay Giesecke, from Stanford's Advanced Financial Technologies Lab, who will discuss Towards Explainable AI: Significance Tests for Neural Networks; Yange Leng, from MIT, whose talk will cover Learning strategic interaction from individual action: A game-theoretic approach, and Martin Haugh, from Imperial College, whose workshop is titled How to Play Fantasy Sports Strategically (and Win).
The afternoon workshops include Peter Decrem, from Citi, with a session on Using AI Machine Learning to Explore Large Streaming Financial Data Sets to Improve Market Making; Darren Vengroff, of Two Sigma, talking about Redefining NYC neighborhoods using open data and machine learning; Amanda Stent, of Bloomberg, discussing Text Analytics in Finance; and Rama Cont, of Oxford, whose talk is titled Forecasting price moves from order flow: perspectives from Deep Learning. The day will end with a reception in the lobby of Lerner Hall.
The Financial Analytics Center is one of the key centers at the Data Science Institute. It brings together expertise in finance theory, machine learning, statistics, signal processing, operations, and natural language processing, and supports collaborations with an appropriately trained student body as well as with the financial industry. The Center for Financial Engineering is an interdisciplinary research center established in 2007 to encourage research on financial engineering and mathematical modeling in finance. It aims at fostering research collaboration among Columbia faculty, graduate students and affiliates, facilitating their contacts with financial institutions and corporations and enhancing the visibility of Columbia as a center for research and innovation in financial engineering.
–By Robert Florida
Posted:May 13 2019