World Business StrategiesServing the Global Financial Community since 2000

Friday 18th October

08.30 - 09.00
Morning Welcome Coffee
Stream Chair:

Chair:

Emmanuel Lesur:

General Manager, Global Sales in Fintech Software Solutions, Yields.io

Emmanuel Lesur: General Manager, Global Sales in Fintech Software Solutions, Yields.io

09.00 - 09.45
Machine Learning & Quantum Computing Techniques Stream
Robust Bayesian SUR for Model Benchmarking
  • Model dependencies
  • Data quality, representativeness, preprocessing, controls
  • Framework and assumptions
  • Model design and performance testing
  • Model selection, backtesting, benchmarking, sensitivity testing, model uncertainty
  • Model monitoring
  • Limitations

Chamberlain Mbah:

Head of AI and Machine Learning, Yields.io

Chamberlain Mbah: Head of AI and Machine Learning, Yields.io

He is also a visiting lecturer at the African Institute of Mathematical Sciences (AIMS) Senegal, where he introduces concepts of machine learning/statistics to the students.
Chamberlain holds a Bachelor and a Master degree in Mathematics and Computer Sciences at University of Buea Cameroon. Later he obtained a Masters degree and a Ph.D. in Statistical Data Analysis at the University of Ghent, Belgium.”

09.45 - 10.30
Machine Learning & Quantum Computing Techniques Stream
“Asymptotics control in the Neural Network”
  • Kolmogorov-Arnold Theorem and asymptotics
  • Control variate functions with the right asymptotics
  • Neural networks with zero asymptotics
  • Numerical experiments

A. Antonov and V. Piterbarg

Alexandre Antonov:

Chief Analyst, Danske Bank

Alexandre Antonov: Chief Analyst, Danske Bank

Alexandre Antonov received his PhD degree from the Landau Institute for Theoretical Physics in 1997. He worked for Numerix during 1998-2017 and recently he has joined Danske Bank as the Chief Analyst in Copenhagen.

His activity is concentrated on modeling and numerical methods for interest rates, cross currency, hybrid, credit and CVA/FVA/MVA. AA is a published author for multiple publications in mathematical finance and a frequent speaker at financial conferences.

He has received a Quant of Year Award of Risk magazine in 2016.

10.30 - 11.00
Morning Break and Networking Opportunities
11.00 - 11.45
Machine Learning & Quantum Computing Techniques Stream
P Pricing by (Q learning and) Regression
  • Consider an instantly but not intertemporally arbitrage-free P-measure market of stock and its options.  Options here are “delta one” first class tradeables.
  • Parameterize dynamics of the option surface under P-measure in terms of non-parametric (piecewise constant) Dupire local volatility.
  • Jointly model the local volatility levels as a VAR(MA) process. This guarantees absence of instantaneous arbitrage: at every future moment option prices are non-arbitrageable.
  • Price non-vanilla options by an AMC-like recursion/regression for the value function and quadratically optimal hedging strategy.  Show how limiting cases are reproduced under P measure.

Andrey Chirikhin:

Founder, Quantitative Recipes

Andrey Chirikhin: Founder, Quantitative Recipes

Andrey was formerly Head of Modelling and Quantitative Analytics for L1 Treasury, part of a USD 25bn privately held investment vehicle LetterOne. Prior to LetterOne, Andrey was MD and Head of CVA and CCR quantitative Analytics at RBS. There he has created and run the front office cross asset CVA quant team. He also restructured and led the risk-side quant team charged with delivering a new Basel III compliant internal CCR methodology. The system utilizing the newly delivered methodology has won the 2013 Internal System of the year Risk award. In his 20 year career in investment banking, Andrey held several leadership and senior quant positions at Goldman Sachs, HSBC and Dresdner Kleinwort. Andrey Chirikhin holds PhD in Theoretical Statistics from Warwick University (UK), MBA from INSDEAD and MSc in Applied Mathematics from Moscow Institute for Physics and Technology (Phystech).

Since 2018 Andrey runs his own company, Quantitative Recipes, that advises on wide rage of XVA, long-term market modelling for risk and quant infrastructure.

11.45 - 12.30
Machine Learning & Quantum Computing Techniques Stream
Validating / Auditing ML Models

Machine Learning Models are like any other models, but different. You probably had presentation or experiences on how to use them in order to get good value out of their relative complexity.

This presentation will be about what, from a Model Audit point of view, should be set in place in order to avoid them failing despite all the efforts you put in building them.

But also incidentally how some can exploit their specific vulnerability to make them fail.

It will cover :

  • Model Risk : Machine Learning models are like any other model.
    • So what’s required for non-ML models?
  • Artificial Intelligence and ML : Machine Learning models are not exactly like any other model.
    • What could make them fail ?
    • How could they fail
  • Tricks or treats? Application to Neuron networks or Random Forest

Gilles Artaud: 

Head of Model Internal Audit, Group Crédit Agricole

Gilles Artaud: Head of Model Internal Audit, Group Crédit Agricole

Gilles Artaud has been working in investment banking for the last 20 years, where he held various positions within Quant, Front Office and Risk Department, working all along on many underlying types, pricing, validation, regulatory and economic capital, market risk and counterparty credit risk topics.

After setting in place the methodology and library for CCR and CVA, he lead XVA, initial margins on non-cleared transactions, and many regulatory topics.

His current “hot” topics are XVAs (CVA DVA FVA AVA MVA…) and impact of new regulatory requirements on derivatives, among which SA-CCR, NSFR, FRTB and FRTB-CVA and Artificial Intelligence technologies in Risk Management.

12.30 - 13.30
Lunch
13.30 - 15.00
Machine Learning & Quantum Computing Techniques Stream
Extended Talk: Deep Analytics

Abstract:

We apply deep learning to resolve the conundrum of revaluation of large, diverse trading books in the context of regulatory simulations and also offer a solution to MVA.

Antoine Savine:

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

15.00 - 15.15
Afternoon Break and Networking Opportunities
15.15 - 16.00
All Streams
Closing Presentation: NLP and Quant Investing: Finding Signals in the Noise
At it’s most basic, Natural Language Processing can be seen as a way for a computer to understand human language. Given that the vast majority of data comes in unstructured form, the potential opportunities for structuring, modeling and implementing text-based investment strategies are huge. For all those on the buy-side, understanding NLP – and the data, tools and processes needed to make it a success – is essential.
  • How NLP has developed alongside the explosion of text-based content.
  • Use-cases today for NLP within investment processes.
  • Factors to consider when building NLP into a quantitative strategy

Saeed Amen

Founder: Cuemacro

Saeed Amen: Founder: Cuemacro

Saeed has a decade of experience creating and successfully running systematic trading models at Lehman Brothers and Nomura. He is the founder of Cuemacro, Cuemacro is a company focused on understanding macro markets from a quantitative perspective. He is the author of ‘Trading Thalesians – What the ancient world can teach us about trading today’ (Palgrave Macmillan), and graduated with a first class honours master’s degree from Imperial College in Mathematics& Computer Science.

Friday 18th October

08.30 - 09.00
Morning Welcome Coffee
Stream Chairs:

Helyette Geman:

Professor of Mathematical Finance, Birkbeck – University of London & Johns Hopkins

Helyette Geman, PhD, PhD: Professor of Mathematical Finance, Birkbeck – University of London & Johns Hopkins

Helyette GEMAN is a Professor of Mathematical Finance at Birkbeck – University of London and at Johns Hopkins University. She is a Graduate of Ecole Normale Supérieure in Mathematics, holds a Masters degree in Theoretical Physics, a PhD in Probability from the University Pierre et Marie Curie and a PhD in Finance from the University Pantheon Sorbonne.
She has been a scientific advisor to a number of major energy and mining companies for the last 20 years, covering the trading of crude oil, natural gas, electricity as well as metals in companies such as EDF Trading, Louis Dreyfus or BHP Billiton and was named in 2004 in the Hall of Fame of Energy Risk.
Prof Geman was previously the head of Research and Development at Caisse des Depots. She has published more than 140 papers in major finance journals including the Journal of Finance, Mathematical Finance, Journal of Financial Economics, Journal of Banking and Finance and Journal of Business. She has also written the book entitled Insurance and Weather Derivatives and is a Member of Honor of the French Society of Actuaries.
Her research includes exotic option pricing for which she got the first prize of the Merrill Lynch awards, asset price modeling through the introduction of transaction time (JOF, 2000); she is one of the authors of the CGMY pure jump Levy model (2002). Prof Geman had organized in 2000 at College de France the first meeting of the Bachelier Finance Society, with Paul Samuelson, Robert Merton and Henry McKean as keynote speakers.
Her book, ‘Commodities and Commodity Derivatives’ is the reference in the field. She was a Scientific Expert on Agriculture for the European Commission and is on the Board of the Bloomberg Commodity Index.
She counts among her numerous PhD students Nassim Taleb, author of the Black Swan

Rita Laura D’Ecclesia

Professor: Università degli Studi di Roma “La Sapienza”

Rita Laura D’Ecclesia Professor: Università degli Studi di Roma “La Sapienza”

09.00 - 09.45
Volatility & Modelling Techniques Stream
Smart Derivative Contracts: Rethinking OTC Derivatives in the Digital Era

We define a financial derivative contract whose entire life-cycle can be processed in a fully digital and algorithmic way making use of modern distributed ledger technologies. By rethinking life-cycle events and contract states of classical OTC derivatives, the full automation of the trade life-cycle is achieved and contractual risks inherent to current derivatives are eliminated by design.

The concept of the smart derivative contract has the potential to disrupt the current architecture of the OTC derivatives market built around central clearing counterparts.

Rebecca Declara:

Interest Rates Options Trader, BayernLB

Rebecca Declara: Interest Rates Options Trader, BayernLB

Rebecca is a quantitative finance professional, experienced both in trading and risk modelling. She complements her current role as an interest rate options trader at BayernLB with quantitative projects focusing on the transparent pricing and risk management of structured interest rate derivatives. Prior to this role, Rebecca worked as an XVA Quant where she contributed to the development and implementation of several methodological XVA initiatives (e.g. MVA) and was the responsible project leader for the implementation of a bank-wide XVA pricing and risk management system. Rebecca holds a M.Sc. in Financial Mathematics and a M.A. in Philosophy from the Ludwig Maximilian University of Munich.

09.45 - 10.30
Volatility & Modelling Techniques Stream
Efficient Numerical Techniques for Parametric Problems in Option Pricing

This talk presents some recent developments of efficient numerical techniques: The implied volatility is one of the most frequently used functions in finance. Its efficient computation for instance supports large-scale computations in machine learning algorithms to predict the volatility. In (Glau, Herold, Madan, and Pötz 2017) https://arxiv.org/abs/1710.01797, a new method to efficiently compute the implied volatility based on multivariate Chebyshev interpolation was introduced. We also show how further adaptations of the algorithm have lead to an even higher performance.

The method relies on polynomial interpolation. It is well-known that polynomial interpolation can be very efficient for low dimensional problems. Can we make polynomial interpolation also efficient for high dimensional problems?

To this end propose in (Glau, Kressner, Statti 2019) https://arxiv.org/abs/1902.04367 a methodology combining low-rank tensor completion and Chebyshev interpolation.

Kathrin Glau:

Lecturer in Financial Mathematics, Queen Mary University of London

Kathrin Glau: Lecturer in Financial Mathematics, Queen Mary University of London

Kathrin Glau currently is a Lecturer in Financial Mathematics at Queen Mary University of London & FELLOW co-founded by Marie Skłodowska Curie at École Polytechnique Fédérale de Lausanne. Between 2011 and 2017 she was Junior Professor at the Technical University of Munich. Prior to this she worked as a postdoctoral university assistant at the chair of Prof. Walter Schachermayer at the University of Vienna. In September 2010 she completed her Ph.D. on the topic of Feynman-Kac representations for option pricing in Lévy models at the chair of Ernst Eberlein.

Her research is driven by the interdisciplinary nature of computational finance and reaches across the borders of finance, stochastic analysis and numerical analysis. At the core of her current research is the design and implementation of complexity reduction techniques for finance. Key to her approach is the decomposition of algorithms in an offline phase, which is a learning step, and a fast and accurate online phase. The methods range from model order reduction of parametric partial differential equations to learning algorithms and are designed to facilitate such diverse tasks as uncertainty quantification and calibration, real-time pricing, real-time risk monitoring, and intra-day stress testing.

10.30 - 11.00
Morning Break and Networking Opportunities
11.00 - 11.45
Volatility & Modelling Techniques Stream
Futures and Options on Bitcoins: A Tentative Arbitrage Approach

Helyette Geman:

Professor of Mathematical Finance, Birkbeck – University of London & Johns Hopkins

Helyette Geman, PhD, PhD: Professor of Mathematical Finance, Birkbeck – University of London & Johns Hopkins

Helyette GEMAN is a Professor of Mathematical Finance at Birkbeck – University of London and at Johns Hopkins University. She is a Graduate of Ecole Normale Supérieure in Mathematics, holds a Masters degree in Theoretical Physics, a PhD in Probability from the University Pierre et Marie Curie and a PhD in Finance from the University Pantheon Sorbonne.
She has been a scientific advisor to a number of major energy and mining companies for the last 20 years, covering the trading of crude oil, natural gas, electricity as well as metals in companies such as EDF Trading, Louis Dreyfus or BHP Billiton and was named in 2004 in the Hall of Fame of Energy Risk.
Prof Geman was previously the head of Research and Development at Caisse des Depots. She has published more than 140 papers in major finance journals including the Journal of Finance, Mathematical Finance, Journal of Financial Economics, Journal of Banking and Finance and Journal of Business. She has also written the book entitled Insurance and Weather Derivatives and is a Member of Honor of the French Society of Actuaries.
Her research includes exotic option pricing for which she got the first prize of the Merrill Lynch awards, asset price modeling through the introduction of transaction time (JOF, 2000); she is one of the authors of the CGMY pure jump Levy model (2002). Prof Geman had organized in 2000 at College de France the first meeting of the Bachelier Finance Society, with Paul Samuelson, Robert Merton and Henry McKean as keynote speakers.
Her book, ‘Commodities and Commodity Derivatives’ is the reference in the field. She was a Scientific Expert on Agriculture for the European Commission and is on the Board of the Bloomberg Commodity Index.
She counts among her numerous PhD students Nassim Taleb, author of the Black Swan

11.45 - 12.30
Volatility & Modelling Techniques Stream
Do Diamonds Shine in Investor Portfolios

Diamonds are emerging as a new investment asset, providing great opportunities for trading, investing and diversification. Hedge funds and financial intermediaries have shown increased interest in the market and recent available data allow us to study its features and dynamics. The lack of a standardization system for the diamond commodity prevented the existence of an exchange regulated trading platform for diamonds which is being created and is starting to play an important role. Over the last decade trading diamonds has been advertised by banks and financial intermediaries as a hedge even if not enough evidence was provided. Diamond stocks have also been considered as a promising diversification asset for investors’ portfolios (McKeough, 2015; Neil, 2014; Wilson and England, 2014; Cameron, 2014), though, to our best knowledge, neither academic scholars nor industry professionals have tested this hypothesis.
In this paper we test if diamonds represent an hedge or safe haven in the investment worlds and if diamond stocks represent an alternative to investing in diamonds. We address two practical investment questions: Can an investment in diamonds represent a hedge for an investment portfolio? Is diamond equity sensitive to diamond prices? We use Polished Prices data set to build a Diamond basket index (D’Ecclesia Jotanovic 2017) and the diamond mining stock prices traded at the main stock markets to investigate the safe haven or hedge hypothesis and to study the relationship existing between diamond prices ad diamond stocks. Our results show that Diamonds can represent a hedge in investor’s portfolios, however the market of diamond-mining stocks does not represent a valid investment alternative to the diamond commodity market.

Rita Laura D’Ecclesia

Professor: Università degli Studi di Roma “La Sapienza”

Rita Laura D’Ecclesia Professor: Università degli Studi di Roma “La Sapienza”

12.30 - 13.30
Lunch
13.30 - 15.00
Volatility & Modelling Techniques Stream
Extended Talk: Default Timing and Correlation Model for DRC (FRTB Internal Model)

Part 1 : Default timing / Correlation model (F.Bergeaud)

  • Calibrating Asset and Default correlation
  • Correlated default timing in structural model
  • Hedging extreme losses

Part 2 : Model risk in DRC: Choice of Copula (M.Predescu)

  • Choice of Copula
  • Copula Estimation
  • Impact on DRC

Francois Bergeaud:

FRTB Lead Quantitative Analyst, BNP Paribas

Francois Bergeaud: FRTB Lead Quantitative Analyst, BNP Paribas

FRTB Lead Quantitative Analyst at BNP Paribas

Previously , Head of the XVA quantitative analytics, Modelling and development of the XVA engine and analytics used globally by the XVA desk and corporate sales.

Quantitative Analysis / Library development / Solution, XVA , Interest Rates, Fixed Income, Credit Derivatives, Inflation, FX, DRC, FRTB, PhD Mathematics ECP and Courant Institute (NYU) ; Graduated from Ecole Centrale Paris

Mirela Predescu:

Deputy Head of Credit – Market and Counterparty Risk, BNP Paribas

Mirela Predescu, Deputy Head of Credit – Market and Counterparty Risk, BNP Paribas

Mirela Predescu is a manager at BNP Paribas, Risk Analytics & Modelling in London.  Mirela is also a Visiting Lecturer at Cass Business School, City University London. Prior to BNP Paribas, Mirela has held positions in the portfolio modelling team at Lloyds Banking Group and the quantitative analytics team at Fitch Solutions. Before moving to the financial industry, Mirela was a University Lecturer at Saïd Business School, University of Oxford. Mirela holds a PhD in Finance from Rotman School of Management, University of Toronto and an MA in Economics from University of Toronto.

15.00 - 15.15
Afternoon Break and Networking Opportunities
15.15 - 16.00
All Streams
Closing Presentation: NLP and Quant Investing: Finding Signals in the Noise
At it’s most basic, Natural Language Processing can be seen as a way for a computer to understand human language. Given that the vast majority of data comes in unstructured form, the potential opportunities for structuring, modeling and implementing text-based investment strategies are huge. For all those on the buy-side, understanding NLP – and the data, tools and processes needed to make it a success – is essential.
  • How NLP has developed alongside the explosion of text-based content.
  • Use-cases today for NLP within investment processes.
  • Factors to consider when building NLP into a quantitative strategy

Saeed Amen

Founder: Cuemacro

Saeed Amen: Founder: Cuemacro

Saeed has a decade of experience creating and successfully running systematic trading models at Lehman Brothers and Nomura. He is the founder of Cuemacro, Cuemacro is a company focused on understanding macro markets from a quantitative perspective. He is the author of ‘Trading Thalesians – What the ancient world can teach us about trading today’ (Palgrave Macmillan), and graduated with a first class honours master’s degree from Imperial College in Mathematics& Computer Science.

Friday 18th October

08.30 - 09.00
Morning Welcome Coffee
Stream Chair:

Ignacio Ruiz:

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.

Ignacio has several publications in the space of quantitative risk management and pricing. He has also published a comprehensive guide to the subject of XVA Desks and Risk Management.

He holds a PhD in nano-physics from Cambridge University.

09.00 - 09.45
XVA, AAD, MVA & Initial Margin Stream
Applying Machine Learning for Troubleshooting CVA Exposure Calculation
  • Applying convolutional neural network to characterizing and troubleshooting CVA exposures used in XVA and Risk.
  • How we choose the model specification to strike a balance between model performance and decision speed.
  • Compare the model performance with human analyst.
  • Possible extension for this model to other area like FRTB.

Shengyao Zhu:

Senior Quantitative Analyst, XVA Trading Desk, Nordea

Shengyao Zhu: Senior Quantitative Analyst, XVA Trading Desk, Nordea

Shengyao currently works as a senior quantitative analyst at Nordea XVA trading desk. Before this, Shengyao worked in different banks in Europe and Asia as quantitative analyst for market risk and counterparty credit risk.  Shengyao hold a master degree of mathematical modelling from Technical University of Denmark and a Bachelor degree from Central University of Finance and Economics, Beijing China.

09.45 - 10.30
XVA, AAD, MVA & Initial Margin Stream
KVA Under IMM and Advanced Approaches

The two largest components of Capital Valuation Adjustment (KVA) are the costs of Counterparty Credit Risk (CCR) and CVA capital. For a bank using the most advanced capital models – Internal Models Method for CCR and the incoming SA-CVA capital –an accurate KVA involves forward simulating expected exposures (EE) over the lifetime of the portfolio – potentially a Monte Carlo in a Monte Carlo. We present a practical regression-based solution.

  • Simulating EE: from regulatory stressed real-world measure to market implied measure
  • A comparative study of regression vs brute force nested Monte Carlo
  • SA-CVA: extending from simulating forward EE to simulating forward CVA sensitivities

Justin Chan:

Quantitative Strategy, Adaptiv, FIS

Justin Chan: Quantitative Strategy, Adaptiv, FIS

Justin Chan has over 11 years of experience in financial risk management and capital markets. Mr. Chan has a deep focus on quantitative modelling in areas such as xVA, credit exposure, and collateral simulations. He is currently responsible for the Risk Quantitative Strategy and Innovation program at FIS. Prior to FIS, Mr. Chan worked at Manulife Financial as a manager in corporate risk management.

Mr. Chan studied engineering science (BASc), and theoretical physics (MSc) at University of Toronto, where he also holds a Master of Mathematical Finance (MMF) degree.

10.30 - 11.00
Morning Break and Networking Opportunities
11.00 - 11.45
XVA, AAD, MVA & Initial Margin Stream
Efficient Calculation Techniques for Credit Exposure in the Presence of Initial Margin
  • Modeling collateralized exposure
  • Producing exposure on a daily simulation time grid without daily revaluations or daily IM calculations
  • Reducing simulation noise in the presence of IM
  • Alternatives to calculating IM along simulation paths

Michael Pykhtin:

Manager, Quantitative Risk, U.S. Federal Reserve Board

Michael Pykhtin: Manager, Quantitative Risk, U.S. Federal Reserve Board

Michael Pykhtin is a manager in the Quantitative Risk section at the U.S. Federal Reserve Board. Prior to joining the Board in 2009 as a senior economist, he had a successful nine-year career as a quantitative researcher at Bank of America and KeyCorp. Michael has edited “Counterparty Risk Management” (Risk Books, 2014) and “Counterparty Credit Risk Modelling” (Risk Books, 2005). He is also a contributing author to several recent edited collections. Michael has published extensively in the leading industry journals; he has been an Associate Editor of the Journal of Credit Risk since 2007. Michael is a two-time recipient of Risk Magazine’s Quant of the Year award (for 2014 and 2018). Michael holds a Ph.D. degree in Physics from the University of Pennsylvania and an M.S. degree in Physics and Applied Mathematics from Moscow Institute of Physics and Technology.

11.45 - 12.30
XVA, AAD, MVA & Initial Margin Stream
Run-time AAD-Compiler

Here, we present a new approach for Automatic Adjoint Differentiation with a special focus on computations where derivatives dF(X)/dX are required for multiple instances of vectors X. In practice, the presented approach is able to calculate all the differentials faster than the primal (original) C++ program for F() and achieve a major performance breakthrough in the AAD field. Our innovative idea is to use the Overloaded Operators to auto-generate AD-version of the primal function at run-time. The created AD-functions can be used for different X[i] instead of performing classic OO AAD approach on each F(X[i])..

The produced functions have multiple advantages:

  • Eliminating Operator Overloading overhead completely
  • Full utilization of native CPU vectorization
  • Segregating simulation data from quant library and enabling safe multi-thread simulations even if underlying program isn’t multi-thread safe.

The just-in-time AAD-Compiler by MatLogica is a completely enabled OO AAD library built around this approach and offers the following features:

  • Optimized machine binary code generation for both runtime performance and quick code generation
  • Streaming compilation
  • Proprietary AD code-folding compression
  • Support for complete AVX2 and AVX512 vectorization

Dmitri Goloubentsev:

Head of Automatic Adjoint Differentiation, Matlogica

Dmitri Goloubentsev: Head of Automatic Adjoint Differentiation, Matlogica

Dmitri has 15 years of combined experience in model development working on C++ quant libraries. He worked as a Senior Quant Analyst in interest rate derivatives and played a leading role in delivering XVA solution at a major Canadian bank. Prior to focusing on AAD, he was responsible for construction of SIMM/MVA model. Dmitri earned his degree in Maths and Applied Maths from the Moscow State University.

12.30 - 13.30
Lunch
13.30 - 15.00
XVA, AAD, MVA & Initial Margin Stream
Extended Talk: Balance Sheet XVA by Deep Learning and GPU

(joint work S Crépey, Univ Evry, France, and Rodney Hoskinson, ANZ Bank, Singapore)

ABSTRACT:

Two competing XVA paradigms are a semi-replication framework and a cost-of-capital, incomplete market approach. Burgard and Kjaer once dismissed an earlier incarnation of the Albanese and Crépey holistic, incomplete market XVA model as being elegant but difficult to solve explicitly. We show that the model (set on a forward/backward SDE formulation) is not only elegant, but also able to be solved efficiently using GPU computing combined with AI methods in a whole bank balance sheet context. We calculate the Mark-to-Market process cube (or its increment, in the context of trade incremental XVA computations) using GPU computing and the XVA process cube using Deep Learning (including joint ES and VaR) Regression methods.

Stéphane Crépey:

Professor of Mathematics, University Of Evry

Stéphane Crépey: Professor of Mathematics, University Of Evry

Stéphane Crépey is professor at the Mathematics Department of University of Evry (France), head of Probability and Mathematical Finance and head of the Engineering and Finance branch (M2IF) of the Paris-Saclay Master Program in Financial Mathematics. His research interests are financial modeling, counterparty and credit risk, numerical finance, as well as related mathematical topics in the fields of backward stochastic differential equations and partial differential equations. He is the author of numerous research papers and two books: “Financial Modeling: A Backward Stochastic Differential Equations Perspective” (S. Crépey, Springer Finance Textbook Series, 2013) and “Counterparty Risk and Funding, a Tale of Two Puzzles” (S. Crépey, T. Bielecki and D. Brigo, Chapman & Hall/CRC Financial Mathematics Series, 2014).

He is an associate editor of SIAM Journal on Financial Mathematics, International Journal of Theoretical and Applied Finance, and a member of the scientific council of the French financial markets authority (AMF). Stéphane Crépey graduated from ENSAE and he holds a PhD in applied mathematics from Ecole Polytechnique and INRIA Sophia Antipolis.

Rodney Hoskinson:

Director, Quantitative Support (Strategic Trading and Funding), ANZ Banking Group

Rodney Hoskinson: Director, Quantitative Support (Strategic Trading and Funding), ANZ Banking Group

Rodney Hoskinson is a Director at ANZ Banking Group having joined in 2017 in the front office quantitative team for the Group’s Global Markets business. In this role based in Singapore, he is responsible for XVA support, primarily focussed on the strategic and quantitative development of the KVA (capital valuation adjustment). Before joining ANZ he was manager, KVA desk quantitative analysis in Fixed Income, Currencies and Commodities at National Australia Bank. His past experience includes a Director role at PwC Australia focussed on financial services consulting and audit support in market risk, implementation of economic capital models for several large insurers, and senior actuarial management roles at QBE Insurance Group. Rodney holds a PhD in Finance from France’s EDHEC Business School and is also a Chartered Enterprise Risk Actuary.

15.00 - 15.15
Afternoon Break and Networking Opportunities
15.15 - 16.00
All Streams
Closing Presentation: NLP and Quant Investing: Finding Signals in the Noise

At it’s most basic, Natural Language Processing can be seen as a way for a computer to understand human language. Given that the vast majority of data comes in unstructured form, the potential opportunities for structuring, modeling and implementing text-based investment strategies are huge. For all those on the buy-side, understanding NLP – and the data, tools and processes needed to make it a success – is essential.

  • How NLP has developed alongside the explosion of text-based content.
  • Use-cases today for NLP within investment processes.
  • Factors to consider when building NLP into a quantitative strategy

Saeed Amen

Founder: Cuemacro

Saeed Amen: Founder: Cuemacro

Saeed has a decade of experience creating and successfully running systematic trading models at Lehman Brothers and Nomura. He is the founder of Cuemacro, Cuemacro is a company focused on understanding macro markets from a quantitative perspective. He is the author of ‘Trading Thalesians – What the ancient world can teach us about trading today’ (Palgrave Macmillan), and graduated with a first class honours master’s degree from Imperial College in Mathematics& Computer Science.

  • Discount Structure
  • Special Offer
    When two colleagues attend the 3rd goes free!

  • Conference + Workshop
    £300 Discount

  • 70% Academic Discount
    (FULL-TIME Students Only)

Event Email Reminder

Error