Director, Portfolio Analytics Bank of America Merrill Lynch
Cristian Homescu: Director, Portfolio Analytics Bank of America Merrill Lynch
Cristian is part of the Portfolio Analytics team within Chief Investment Office, Global Wealth and Investment Management division Bank of America Merrill Lynch. He is developing and investigating quantitative solutions in areas such as investment strategies, goals-based wealth management, asset allocation, machine learning and big data analysis, factor-based investing and risk factor models, portfolio risk and attribution, stress testing and scenario construction. He is very interested in application of state-of-the-art algorithms and numerical methods in wealth and investment management, and in high-performance computing. Prior to joining Bank of America Merrill Lynch, Cristian was a front office quant for Wachovia and Wells Fargo. After supporting interest rate trading desk, he was the lead quant for FX and Commodities trading desks. He has a PhD from Florida State University in computational and applied mathematics, and MSc degrees from University of Paris XI and University of Craiova.
Abstract: Artificial intelligence (AI) algorithms typically operate in high-dimensional spaces, are trained on vast and often heterogeneous data sets, and involve non-linear models that attempt to capture previously unknown patterns. In areas where the results of such algorithms are filtered through human decisions, insight into the pattern generation processes is key to fully capitalizing on the power of AI. Actionable AI (AAI) aims to build trust between collaborating human and software agents. In Finance, AAI is paramount not only to addressing regulatory requirements for transparency, but also to the widespread adoption of AI methods in areas largely driven by human decisions. In this presentation, we delve into the challenges and state-of-the-art solutions for AAI in the context of quantitative modeling for investment management decisions.
Ioana Boier: Independent
I have a Ph.D. in Computer Science from Purdue University. In addition, I have completed graduate coursework in Financial Mathematics at NYU and Big Data at Harvard University. Prior to joining Citadel, I was a Director in the Global Markets Division at BNP Paribas where I managed the Interest Rate Options & Inflation quantitative research team. Before transitioning into Finance, I was a research staff member at the IBM T. J. Watson Research Center.
CEO, Rebellion Research – The Ai Machine Learning Robo Advisor
Alexander Fleiss: CEO, Rebellion Research – The Ai Machine Learning Robo Advisor
Richard V. Rothenberg:
Global AI Corporation & Research Affiliate, Lawrence Berkeley National Laboratory
Richard V. Rothenberg: Executive Director, Global AI Corporation, New York, NY and Research Affiliate, Lawrence Berkeley National Laboratory, Berkeley, CA
CEO Of The Python Quants
Yves Hilpisch: CEO Of The Python Quants
An investigation of how aggregate measures of news article sentiment can prefigure changes in market volatility. This will use a variety of natural language processing techniques – including Sentiment Analysis and doc2vec – to determine aggregate measures of sentiment for an individual stock. The measures are used as an input to a neural network classifier to predict whether the article will lead to market changes.
Product Manager, Suite, LLC
James Baker: Product Manager, Suite, LLC
James Baker is the Product Manager at Suite, LLC, the provider of the ALib fixed income analytics library. James has worked in quantitative, risk and IT roles, at both banks and vendors, in New York, London and Paris. He has a Computer Science Masters from London’s Imperial College, where he focused on Artificial Intelligence. His recent work on Machine Learning has investigated the scope for uncovering useful insights from news reports.
An introduction of computing has revolutionized the financial trading industry, making it possible to track market changes and execute an enormous number of orders at a speed of light. Today, AI-driven trading systems are launching the next wave of innovation that will result in the most significant transformation in finance history. In this talk, Vladyslav Ivanov is going to introduce the audience to the development and production level algorithmic trading pipelines and their primary components. The talk will cover how the three primary subsets of Artificial Intelligence: Machine Learning, NLP, and Vision are being used in an effort to gain better market insights and ultimately a trading advantage. Topics as fetching and storing the time-series financial data, application of alternative data, and aspects of the financial markets will also be discussed. Finally, we will conclude by talking about using AI in conjunction with traditional models for describing market volatility, such as GARCH.
Quantitative Researcher, Outremont Technologies
Vladyslav Ivanov: Quantitative Researcher, Outremont Technologies
Vladyslav Ivanov is a Quantitative Researcher with proven experience in leading systematic trading strategies research and applying statistical learning to problems in quantitative finance. Prior to joining Outremont Technologies, Vladyslav worked at a Chicago Proprietary Trading firm, where he conducted alternative data strategies research and was a product owner of the research framework. He also worked in Quantitative Research at a leading New York Hedge Fund, where he designed and implemented a market regimes analysis system, collaborated with the portfolio manager on alpha strategies, and built large-scale data processing systems.
Vladyslav holds a Bachelor’s degree in Financial Economics with a sequence in Data Science from Claremont McKenna College.
A suitably ICA-corrected Latent Semantic Analysis (pICA) is shown to produce linear factors on text data which are maximally parsimonious according to a specific criterion. The model produces stable and mostly interpretable unsupervised factors for Bloomberg Story-Level data for both News and Twitter feeds. A wide variety of applications of the model are discussed, ranging from factor-specific sentiment aggregation to theme discovery and tracking. Connections to neural net methods are clarified.
Quant and Data Science Research, Bloomberg LP – Adjunct Professor, NYU Courant
Ivailo Dimov: Quant and Data Science Research, Bloomberg LP – Adjunct Professor, NYU Courant Institute
Ivailo Dimov is a senior quant at the Bloomberg L.P. CTO Office, where he provides quantitative and data science solutions to management, external and internal clients. He has worked on both traditional derivative, risk and alpha modeling as well as alternative data research. At Bloomberg, he has led projects on market consensus, broker-algo selection, recommendation systems, automated news and news topic modeling. Ivailo is also an Adjunct Professor at the NYU Courant Institute, where he teaches Data Science in Quantitative Finance.