Tuesday 17th November: Machine Learning in Quantitative Finance
Presenter to be confirmed
Managing Director, Global Head of Credit and Commodities Quantitative Analysis, Citi
Youssef Elouerkhaoui: Managing Director, Global Head of Credit and Commodities Quantitative Analysis, Citi
Youssed Elouerkhaoui is the global Head of Credit Quantitive Analysis at Citi. His group supports all aspects of modelling and product development across desks, thais includes: Flow Credit Trading, Correlation Trading, CDOs, Exotics and Emering Markets.
He also supports CVA, Funding and Regulatory Capital for Credit Markets. Prior to this, he was a Director in the Fixed Income Derivatives Quantitative Research Group at UBS, where he was in charge of developing and implementing models for the Structured Credit Desk. Before joining UBS, Youssef was a Quantitative Research Analyst at Credit Lyonnais supporting the Interest Rates Exotics business. He has also worked as a Senior Consultant in the Risk Analytics and Research Group at Ernst & Young. He is a graduate of Ecole Centrale Paris and he holds a PhD in Mathematics from Paris-Dauphine University.
The computation of risk metrics poses a huge computational challenge to banks. Many different techniques have been developed and implemented in the last few years to try and tackle the problem. We focus on Chebyshev tensors enhanced by machine learning, showing why they are such powerful pricing approximators in risk calculations. We show how the presented unique mix of techniques can be applied in different calculations. We illustrate with Numerical results in Counterparty Credit Risk, Market Risk and Dynamic initial Margin. In particular, We will give special attention on how to side-step the curse of dimensionality and how machine learning techniques can be used to boost Chebyshev tensors.
Founder & CEO, MoCaX Intelligence
Ignacio Ruiz: Founder & CEO, MoCaX Intelligence
Ignacio Ruiz has been the head strategist for Counterparty Credit Risk, exposure measurement, for Credit Suisse, as well as the Head of Risk Methodology, equities, for BNP Paribas. In 2010, Ignacio set up iRuiz Consulting as an independent advisory business in this field. In 2014, Ignacio founded iRuiz Technologies to develop and commercialise MoCaX Intelligence.
He holds a PhD in nano-physics from Cambridge University.
Managing Director and XVA Lead Quant, Scotiabank
Andrew Green: Managing Director and XVA Lead Quant, Scotiabank
Andrew Green is a Managing Director and lead XVA Quant at Scotiabank in London. Prior to joining Scotiabank, Andrew held roles as a quantitative analysis in several different banks in London. He is the author of XVA: Credit, Funding and Capital Valuation Adjustments, published by Wiley.
Differential machine learning (ML) extends supervised learning, with models trained on examples of not only inputs and labels, but also differentials of labels to inputs.
Differential ML is applicable in all situations where high quality first order derivatives wrt training inputs are available. In the context of financial Derivatives risk management, pathwise differentials are efficiently computed with automatic adjoint differentiation (AAD). Differential ML, combined with AAD, provides extremely effective pricing and risk approximations. We can produce fast pricing analytics in models too complex for closed form solutions, extract the risk factors of complex transactions and trading books, and effectively compute risk management metrics like reports across a large number of scenarios, backtesting and simulation of hedge strategies, or capital regulations.
The article focuses on differential deep learning (DL), arguably the strongest application. Standard DL trains neural networks (NN) on punctual examples, whereas differential DL teaches them the shape of the target function, resulting in vastly improved performance, illustrated with a number of numerical examples, both idealized and real world. In the online appendices, we apply differential learning to other ML models, like classic regression or principal component analysis (PCA), with equally remarkable results.
This paper is meant to be read in conjunction with its companion GitHub repo this https URL, where we posted a TensorFlow implementation, tested on Google Colab, along with examples from the article and additional ones. We also posted appendices covering many practical implementation details not covered in the paper, mathematical proofs, application to ML models besides neural networks and extensions necessary for a reliable implementation in production.
Quantitative Research, Danske Bank
Antoine Savine: Quantitative Research, Danske Bank
Antoine Savine has worked for various Investment Banks since 1995, along Bruno Dupire, Leif Andersen and Marek Musiela. He was Global Head of Quantitative Research for Fixed Income, Currency and Credit Derivatives for BNP-Paribas 1999-2009, and currently works in Copenhagen for Danske Bank, where his work with Jesper Andreasen earned the In-House System of the Year 2015 Risk Award. His upcoming publications in Wiley’s Computational Finance series are dedicated to teaching the technologies implemented in those award-winning systems.
Antoine also teaches Volatility Modeling and Numerical Finance in the University of Copenhagen’s Masters of Science in Mathematics-Economics. The curriculum for his Numerical Finance lectures is being published by Wiley under the name “AAD and Parallel Simulations”.
Antoine holds a Masters from the University of Paris (Jussieu) and a PhD from the University of Copenhagen, both in Mathematics.
Brian Norsk Huge:
Chief Quantitative Analyst, Danske Markets
Brian Norsk Huge: Chief Quantitative Analyst, Danske Markets