Quant Summit Europe 2023


Registration and refreshments

08:00 - 08:50

08:5009:00 welcome and macro-outlook - latest trends from risk, what’s next?

08:50 - 09:00

Mauro Cesa

Quantitative finance editor

Mauro Cesa is quantitative finance editor for, based in London. He leads the team responsible for the publication of quantitative research across all brands of the division. The section of he manages, Cutting Edge, publishes peer-reviewed papers on derivatives, asset and risk management, and commodities.
Mauro holds a degree in economics from the university of Trieste and a masters in quant finance from the University of Brescia.


Quant of the year keynote
Keynote presentation

09:00 - 09:35

Vladimir Piterbarg

Managing director, head of quantitative analytics and quantitative development

NatWest Markets

Vladimir Piterbarg is the global head of Quantitative Analytics at NatWest Markets since 2018. He held similar positions at Rokos Capital Management LLP, Barclays Capital/Barclays investment bank, and Bank of America. Vladimir Piterbarg has a PhD in Mathematics (Stochastic Calculus) from the University of Southern California. He serves as an associate editor of the Journal of Computational Finance and the Journal of Investment Strategies. Together with Leif Andersen, Vladimir Piterbarg wrote the authoritative, three-volume set of books “Interest Rate Modelling”. He published multiple papers in various areas of quantitative finance, and won Risk Magazine’s Quant of the Year award twice.


Quant of the year "hall of fame"
Quickfire presentations and panel discussion

09:35 - 10:30

Mathieu Rosenbaum

Professor of finance

Ecole Polytechnique

Mathieu Rosenbaum obtained is Ph.D from University Paris-Estin 2007. After being Assistant Professor at École Polytechnique, he became Professor at University Pierre et Marie Curie (Paris 6) in 2011. He is now full-time professor at Ecole Polytechnique, where he is the at the head of the chair "Analytics and Models for Regulation". He is also in charge, with Nicole El Karoui, Gilles Pagès and Emmanuel Gobet, of the Master program “Probability and Finance”.

His research mainly focuses on statistical finance problems, such as market microstructure modeling or designing statistical procedures for high frequency data and on regulatory issues, especially in the context of high frequency trading. In particular, he is one of the organizers of the conference "Market Microstructure, Confronting Many Viewpoints", which takes place every two years in Paris.

Mathieu Rosenbaum has collaborations with various financial institutions, notably BNP-Paribas since 2004. He also has several editorial activities. He is one of the editors in chief of the journal "Market Microstructure and Liquidity", together with F. Abergel, J.P. Bouchaud, J. Hasbrouck and C.A. Lehalle. Furthermore, he is managing editor for "Quantitative Finance" and associate editor for "Electronic Journal of Statistics", "Journal of Applied Probability", "Mathematics and Financial Economics", "Statistical Inference for Stochastic Processes", "SIAM Journal in Financial Mathematics","Springer Briefs" and "Statistics and Risk Modeling". 

He received the Europlace Award for Best Young Researcher in Finance in 2014 and the European Research Council Grant in 2015.

Oleksiy Kondratyev

Quantitative research & development lead

Abu Dhabi Investment Authority (ADIA)

Alexei is Quantitative Research & Development Lead at ADIA.

Formerly a Managing Director and Global Head of Data Analytics at Standard Chartered Bank, Alexei is responsible for providing data analytics services to Corporate, Commercial and Institutional Banking division of Standard Chartered Bank.

He joined Standard Chartered Bank in 2010 from Barclays Capital where he managed a model development team within Credit Risk Analytics. Prior to joining Barclays Capital in 2004, he was a senior quantitative analyst at Dresdner Bank in Frankfurt.

Alexei holds MSc in Theoretical Nuclear Physics from the University of Kiev and PhD in Mathematical Physics from the Institute for Mathematics, National Academy of Sciences of Ukraine.

He was the recipient of the 2019 Quant of the Year award from Risk magazine.


Morning networking break

10:40 - 11:10



Breakout 1: Pricing, market and liquidity risk and capital optimisation

Signature trading

11:00 - 11:30

Blanka Horvath

Associate professor in mathematical and computational finance

University of Oxford and Researcher

Blanka Horvath is a Lecturer in Financial Mathematics at King’s College London as well as a Honorary Lecturer at Imperial College London and a researcher at The Alan Turing Institute, where she is co-lead of the Machine Learning in Finance theme. Blanka holds a PhD in Financial Mathematics from ETH Zurich, a postgraduate Diplom in pure Mathematics from the University of Bonn and an MSc in Economics from the University of Hong Kong. In her latest research she focusses on non-Markovian models of nancial markets such as Rough Volatility models as well as modern DNN- based market generators. Prior to her position at King’s College, Blanka worked at JP Morgan on the re nements of the Deep Hedging programme and the development of generative market simulation models. Her work on DNN-based calibration of Rough Volatility models was awarded the Rising Star Award 2020 of Risk magazine.

Validation of Internal Expected Shortfall Models: comparing risk predictions and risk realizations

11:30 - 12:00

  • Ridge backtest: unique prudential backtest for ES
  • Unveiling monetary discrepancy between predicted and realized risk
  • New paradigm: managing realized risk, not predicted risk
  • ES/VaR joint elicitability and related model selection methods
  • Potential relevance for FRTB Basel III regulation
Carlo Acerbi

Researcher, fellow and author

Ecole Polytechnique Fédérale de Lausanne

Carlo Acerbi is a quantitative financial risk management researcher and professional, author of relevant contributions in the field of banking regulation (2002 coherent definition of ES ; 2019 ES backtesting), asset management liquidity regulation (2013 MSCI LiquidityMetrics) and stress testing (2016 MSCI ST best practices), among others.

He received a PhD in Theoretical Physics (1998, ISAS-SISSA, Trieste, Italy). He served for leading institutions in the financial risk industry (Banca Intesa, Abaxbank, McKinsey, RiskMetrics-MSCI, Banque Pictet). He teaches Advanced Derivatives at Bocconi University, Milan, Advanced Risk Management Topics at EPFL Lausanne (from Q4 2023) and he's Honorary Professor at Corvinus University, Budapest.

Bridging the gap risk reloaded: a comprehensive methodology for wrong way risk and leveraged exposures

12:00 - 12:30

  • How dangerous is the interplay between WWR, leverage and collateralised exposures?
  • How to generalise MC frameworks in the presence of WWR?
  • And what about non-MC based frameworks such as stress testing and SA-CCR?
Fabrizio Anfuso

Senior technical specialist, PRA

Bank of England

Fabrizio Anfuso, Senior Technical Specialist, PRA, BANK OF ENGLAND

Fabrizio is a leading expert in developing complex risk analytics, quantitative modelling and financial regulations. He has an extensive track-record of heading quant teams onshore and offshore, as well as of taking part in firm-wide programs, such as IMM, BCBS-IOSCO Margin Requirements and IBOR transition. 
In his present and previous roles, Fabrizio has gained a comprehensive knowledge of the full model development cycle, including the model design, the validation of model performance, the IT implementation and the attainment of regulatory compliance.
His main areas of expertise are Counterparty credit risk, Monte Carlo simulations, Internal Models for the trading book (IMM and IMA), derivatives pricing, CCPs & collateral modelling, Initial Margin methodologies and regulatory capital. 
Fabrizio is chairing the master’s course in Counterparty Credit Risk of the ETH / University of Zurich and taught a number of advanced professional trainings in topics such as CCR, capital management and Initial Margin methodologies. 
As part of his academic activities, he has authored numerous research articles in the fields of quantitative finance and condensed matter physics. Fabrizio holds a Ph.D. in Theoretical Physics from the Chalmers University of Technology (Gothenburg, Sweden).



Pricing volatility swaps using Machine Learning

12:30 - 13:00

  • Vol swaps, popular instruments without clear pricing formulas
  •  Deploying machine learning techniques for pricing vol swaps
  • Pure data driven pricing explored and tested
Wim Schoutens


University of Leuven

Wim Schoutens is a quantitative finance professor at the University of Leuven, Belgium.

He has extensive practical experience of model implementation and validation. He is well known for his consulting work with the banking industry and national and supra-national institutions. He is an independent expert advisor to the European Commission, has worked for the IMF and is the author of several books on quantitative finance.

His latest books, co-authored with Dilip Madan, are about the brand new theory of conic finance.
He is also a member of different editorial Boards of international finance journals. Wim is also a founding partner of RiskConcile, a fintech company with roots within the University of Leuven.

He likes arbitrages, politically incorrect statements and making jam.

Networking lunch

13:00 - 14:00

15 Years of Adjoint Algorithmic Differentiation (AAD):  How to better hedge financial risks, crack some of the puzzles of condensed matter and much more with upside-down derivatives

14:00 - 14:30

Following the seminal ‘Smoking Adjoint’ paper by Giles and Glasserman, the development of Adjoint Algorithmic Differentiation (AAD) has revolutionized the way risk is computed in the financial industry. In this talk, I review what makes AAD an important recent innovation in financial risk management and its latest applications, and how the same ideas can be applied in Condensed Matter Physics or in any context in which computing accurately and efficiently a large number of derivatives is beneficial. 

Luca Capriotti

Managing Director - Global Head Quantitative Strategies, Credit

Credit Suisse

Luca Capriotti is a Managing Director at Credit Suisse, based in London, where he works in Quantitative Strategies and he is responsible for Credit Products in Europe, and globally for Corporate Bank and Treasury. Previous to this role, he was US head of Quantitative Strategies Global Credit Products, he has worked in Credit and Commodities Exotics in New York and London and in the cross-asset modeling R&D group of GMAG in the London office.
Luca is also visiting professor at the Department of Mathematics at University College London. His current research interests are in the fields of Machine Learning, Algorithmic Trading, Credit Models and Computational Finance, with a focus on applications of Adjoint Algorithmic Differentiation (AAD) for which he holds a US Patent.

Luca gives regularly gives seminars and courses worldwide. He has served as supervisor and external examiner for Master and PhD programs and as referee for several scientific publications .

Prior to working in Finance, Luca was a researcher at the Kavli Institute for Theoretical Physics, Santa Barbara, California, working in the field of High Temperature Superconductivity and Quantum Monte Carlo methods for Condensed Matter systems.

Luca holds a M.S. cum laude in General Physics from University of Florence (1996), and an M.Phil. and Ph.D. cum laude in Condensed Matter Theory, from the International School for Advanced Studies, Trieste (2000).


Time changes, Fourier transforms and the joint calibration to the S&P500/VIX smiles

14:30 - 15:00

  • We develop a model based on time changed Lévy processes to tackle the 'joint calibration problem', i.e. reproducing the joint S&P500/VIX implied volatility smiles and the VIX futures prices
  • The calibration procedure uses efficient Fourier based pricing schemes.
  • We focus on one specification of the proposed general setting which uses discontinuous processes only.
  • Results show satisfactory performances in solving the joint calibration problem, and therefore confirm that also the class of affine processes can provide a workable fit.
Laura Ballotta

Professor in mathematical finance

Bayes Business School (formerly Cass) 

Laura Ballotta is a Professor of Mathematical Finance at the Faculty of Finance of Bayes Business School (formerly Cass). She works in the areas of quantitative finance and risk management. She has written on topics including stochastic modelling for financial valuation and risk management, numerical methods aimed at supporting financial applications, and the interplay between finance and insurance.

Capital Valuation Adjustments

15:00 - 15:30

This presentation aims to review the rationale for the Capital Valuation Adjustment (KVA):

  • Should derivatives values be adjusted for capital costs? If so in what circumstances and by how much?
  • How does KVA fit in with our understanding of “value”?
  • How does KVA depend on capital levels and how can we determine the return-on-equity?
  • What is the correct KVA for counterparty credit risk?
Matthias Arnsdorf

Global head of counterparty credit risk quantitative research

JP Morgan

Matthias Arnsdorf is a Managing Director and the global head of the Counterparty Credit Risk Quantitative Research team at J.P. Morgan. His responsibilities include the development of credit exposure and funding models for valuation, risk management as well as credit risk capital. He has published and presented numerous articles on credit risk and credit derivatives.

Matthias started his career in finance in 2002 working in credit derivatives quantitative research. Prior to this he spent two years as a post-doctoral researcher at the Niels Bohr Institute in Copenhagen. Matthias holds a PhD in Quantum Gravity from Imperial College London.

Afternoon networking break

15:30 - 16:00

Survival in dark

16:30 - 17:00

Dark venues offer an essential source of liquidity that is important to leverage correctly in execution algorithms. Due to the lack of pre-trade transparency, the liquidity available in such venues is not observable directly and can require estimation from fill data. However, using fill data raises two issues. Firstly, the data is censored by natured and can lead to underestimation if that aspect is not considered. Secondly, the fill data will depend on the choices made by the algorithm, which themselves will typically depend on the estimates of liquidity, thus introducing a feedback loop. We propose an adaptive Bayesian approach inspired by survival analysis to address the first issue in a dynamic way and combine it with a multi-armed bandit solution that addresses the exploration-exploitation dilemma induced by the feedback loop.

Jaya Nagaradjasarma

Head of EMEA equities algo quant research

Bank of America

Jaya Nagaradjasarma is the head of EMEA Equities Algo Quant Research for the Equity eTrading line of business at Bank of America London. She and her team create quantitative solutions that leverage data science, AI, and machine learning to optimise equities execution algorithms.  Before joining Bank of America, Jaya worked in quantitative roles for automated trading desks in several asset classes on the buy side. Her earlier career includes roles at Susquehanna International Group in Dublin, Man AHL in Oxford, and Antares Technologies in Paris. Jaya received her master’s degree from Ecole Nationale de la Statistique et de l'Administration Economique, her DEA in Probability and Finance from Pierre et Marie University (DEA El Karoui), and her Ph.D. in Statistics from the London School of Economics. Jaya currently resides in London.

Damian Bescomes

Vice president, quantitative strategies and data group, EMEA equities execution

Bank of America

Damien Bescombes is a Vice President in the Quantitative Strategies and Data Group for EMEA Equities Execution at BofA Securities. Damien has 6 years of experience designing and building execution algorithms. He also previously worked as a Software Developer at the bank and as a Quant Developer in Energy trading at TotalEnergies. Damien graduated from Supélec in France.

Modelling balance sheet risk post-SVB

17:00 - 17:30

Rama Cont

Chair of mathematical finance, Mathematical Institute

Oxford University

Prof. Rama Cont holds the Chair of Mathematical Finance at Imperial College London and is director of the CFM-Imperial Institute of Quantitative Finance since 2012, after previous appointments at Ecole Polytechnique (France), Columbia University (New York) and Sorbonne (Paris).

His research in finance has focused on modeling of extreme market risks: market discontinuities and breakdowns, liquidity risk, endogenous risk and systemic risk. His 2006 paper on ‘model risk', an early reference on the topic, was the first to propose a quantitative approach to model risk.

Cont has served as a consultant to the BIS, the European Central Bank, the New York Federal Reserve, Norges Bank, the US Commodity Futures Commission (CFTC), the US Office of Financial Research, the IMF and a dozen major CCPs in Europe, Asia, the US and Latin America.

He was awarded the Louis Bachelier Prize by the French Academy of Sciences in 2010 and the Royal Society Award for Excellence in Interdisciplinary Research in 2017 for his research on mathematical modeling in finance.


Pricing derivatives with a variational auto-encoder approach

17:30 - 18:00

  • Motivation: Generative vs Discriminative ML approaches 

  • Mathematical Introduction to Variational Auto-Encoders 

  • Conditional Variational Auto-Encoders 

  • Numerical Implementation 

  • Applications 

Youssef Elouerkhaoui

Managing director, global head of credit and commodities quantitative analysis


Youssef Elouerkhaoui is a Managing Director and the Global Head of Credit and Commodities Quantitative Analysis at Citi. His group supports all modelling and product development activities across businesses. He is also in charge of CVA, Funding and Regulatory Capital for his businesses. Prior to this, he was a Director in the Fixed Income Derivatives Quantitative Research Group at UBS, where he was in charge of model development for the Structured Credit Desk. Before joining UBS, Youssef was a Quantitative Research Analyst at Credit Lyonnais supporting Interest Rates Exotics. He is a graduate of Ecole Centrale Paris and he holds a PhD in Mathematics from Paris-Dauphine University.

Explicit option pricing and volatility surface modelling

18:00 - 18:30

  • fully analytical and empirically sound option pricing formulae exist;
  • they are supported by computationally elementary processes;
  • volatility surfaces can be explicitly determined, and calibration performed analytically
Lorenzo Torricelli

Assistant professor

University of Bologna

Lorenzo is a Lecturer in Mathematics at the University of Bologna. He was previously appointed at the University of Parma and holds a Post-Doc from LMU Munich. Before joining academia, he worked for the Italian Pension Regulator (COVIP) and took part in the EIOPA-Occupational Pension Committees for the revision of the European Union IORP directive on European Pension Funds. 

His research interests are the modelling of the implied volatility surface, applications of jump processes in finance, time changed processes, tempered stable laws, fractional calculus and anomalous diffusions and their applications to finance. He is also interested in economic and mathematical models for public pension systems.

Lorenzo holds an MSc in Mathematics and Finance from Imperial College London and one in Pure Mathematics with a major in Geometry from the University of Roma Tre. His PhD in Mathematics was obtained from UCL London. 


Breakout 2: Data science-driven modelling and computational methods

Cross-Impact in Equity Markets

11:00 - 11:30

The empirical finding that market movements in stock prices may be correlated with the order flow of other stocks has led to the notion of "cross impact" and has prompted the development of multivariate models of market impact. These models are parametrized by a matrix of impact coefficients whose off-diagonal elements are meant to capture how trades in one asset influence the price of other assets, leading to many 'cross impact' parameters which may not be identified solely based on the covariance of returns with order flow. Moreover, empirical evidence suggest that these cross-impact terms are unstable and change sign randomly over time, which poses a problem for their interpretation and use.  
We show that the observed correlations between the returns of an asset and the order flow imbalance (OFI) of other assets have a simpler explanation in terms of common components in order flow across stocks. This commonality in order flow arises naturally from multi-asset trading strategies such as index or ETF portfolios. We provide empirical evidence from order flow and price changes of NASDAQ-100 stocks to support this explanation. Our results show the main determinants of impact to be each stock’s own order flow imbalance (OFI) and the common component of OFI across stocks. Additional ‘cross-impact’ terms account for less than 1% of total impact. This leads to a parsimonious approach for modelling multi-asset impact, which does not require introducing any "cross impact" coefficients. 

Francesco Capponi

Quantitative researcher – trading


Francesco Capponi, PhD, is a member of BlackRock’s Trading Research team.

Francesco is responsible for trading research in both Equities and FX, with an emphasis on market microstructure, transaction cost analysis, and execution research.

Prior to joining BlackRock, he worked at Barclays as a quantitative researcher within the quantitative portfolio construction team.

He holds an MPhil in Economics from the University of Cambridge, and a PhD in Mathematical Finance from Imperial College London.

Machine learning in investment and wealth management: hype versus reality

11:30 - 12:00

The cost of misspecification price impact

12:00 - 12:30

  • What happens when trades are based on an incorrect price impact model? 

  • We derive tractable formulas for these misspecification costs and illustrate them on proprietary trading data

  • The misspecification costs are naturally asymmetric

  • Underestimating impact concavity or impact decay shrinks profits, but overestimating concavity or impact decay can even turn profits into losses

Johanes Muhle-Karbe

Head of mathematical finance

Imperial College

Johannes us the Head of the Mathematical Finance Section at Imperial College London, where he also directs the CFM-Imperial Institute on Quantitative Finance. Before his appointment at Imperial, Johannes held faculty positions at Carnegie Mellon University, the University of Michigan and ETH Zurich. His research focuses on the impact of "market frictions” such as trading costs or asymmetric information on optimal trading strategies.

Rough volatility: fact or artefact?

12:30 - 13:00

  • Is volatility rough? 

  • We explored the question via a model-free approach 

  • Our simulation study suggests roughness of realised vol is not a good indicator of the roughness of Instantaneous vol

  •  This suggests that the origin of the roughness observed in realised volatility lies in the estimation error. 

Purba Das

Lecturer in financial mathematics

Kings College London

Purba Das is a Lecturer in Financial Mathematics at King's College London. Purba's current research interests are Pathwise methods in stochastic analysis and it's application in Finance. Previously, she was a Research Assistant Professor of Mathematics at the University of Michigan (2022-2023). Purba completed her DPhil in Mathematical Institute at the University of Oxford (2018- 2022). Previously Purba received her Master’s, and Bachelor’s degree from Chennai Mathematical Institute, India.

Networking lunch

13:00 - 14:00

Darwinian model risk, reverse stress testing, and HVA

14:00 - 14:30

We acknowledge model risk due to adverse selection in high-priced models for market competitiveness, even if it causes alpha leakage. These losses are initially balanced by gains in hedging, but eventually, financial crises expose model errors, leading to substantial losses during liquidation. This long-term, directional "Darwinian" model risk often eludes traditional risk systems focused on short-term moments beyond order two. Detecting it requires extensive, long-term simulations under extreme scenarios. To manage it, we propose a modified Burnett's hedging valuation adjustment (HVA) as a bridge between global and local valuation models within a bank. Additionally, we advocate for a risk-adjusted reserve, contributing to the bank's KVA alongside HVA. 

Based on joint works with Claudio Albanese, Cyril Benezet, Dounia Essaket, and Stefano Iabichino).

Stephane Crepey

Distinguished professor of mathematics

University of Paris

Stéphane Crépey is Professor of Mathematics at Université Paris Cité, Laboratoire de Probabilités, Statistique et Modélisation (LPSM), Team Mathematical Finance and Numerical Probability. His research interests are counterparty credit risk, XVA analysis, central counterparties, quantitative reverse stress tests; Model risk and uncertainty quantification; Backward stochastic differential equations (reflected, anticipated, in combination with progressive enlargement of filtration,...); Machine learning for finance (learning conditional expectations and risk measures, calibration by neural networks or Gaussian processes, anomaly detection,...). He is the author of numerous research papers published in journals including Annals of Probability, Finance and Stochastics, Mathematical Finance, Stochastic Processes and their Applications or Risk Magazine. He wrote 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 a member of the scientific council of the French financial markets authority (AMF). Stéphane graduated from ENSAE ParisTech and holds a PhD in differential games and mathematical finance from Ecole Polytechnique and INRIA Sophia Antipolis.

Generative modelling for time series and applications to deep hedging

14:30 - 15:00

  • Generative model via Schrödinger Bridge 
  • Financial time series 
  • Kernel regression  
  • Synthetic samples 
  • Deep hedging 
Huyên Pham

Professor of applied mathematics

University of Paris City

Huyên PHAM is a distinguished Professor of Mathematics at Université Paris Cité.  He is also Adjunct Professor at ENSAE and was Chair of Applied Mathematics at the John Von Neumann Institute of VNU-HCM. He leads research in quantitative finance, stochastic analysis and control, machine learning, and is the author of more than 110 publications, including the monograph Continuous time Stochastic Control and Optimization with Financial Applications. He serves on the editorial boards of several international journals and is the co-editor in chief of the journal Applied Mathematics and Optimization. Prof. Pham was appointed member of the Institut Universitaire de France in 2006, awarded the Louis Bachelier prize by the French Academy of Sciences in 2007, and was a plenary speaker at the 9th World congress of the Bachelier Finance Society in 2016, and at the 6th Asian Quantitative Finance Conference in 2018.

Generative AI for Limit Order Book Modelling

15:00 - 15:30

  • We develop a generative model of realistic order flow in financial markets through a first end-to-end model that generates tokenized limit order book (LOB) messages similar to tokenization in large language models (LLM)
  • Out-of-sample results show promising performance in approximating the data distribution, as evidenced by low model perplexity and mid-price returns calculated from the generated order flow exhibit a significant correlation with the data, indicating impressive conditional forecast performance
  • Due to the granularity of generated data, and the accuracy of the model, it offers new application areas for future work beyond forecasting, e.g. acting as a world model in high-frequency financial reinforcement learning applications.
Stefan Zohren

Deputy director

Oxford-Man Institute of Quantitative Finance

Stefan Zohren is deputy director of the Oxford-Man Institute of Quantitative Finance and an associate professor at the Department of Engineering Science at the University of Oxford. He is a fellow of the Turing Institute, the UK’s national institute for artificial intelligence (AI) and data science. Zohren’s research is focused on machine learning in finance, including deep learning, reinforcement learning, network and natural language processing approaches, and early use cases of quantum computing. He works with Man Group on commercial research projects and is a frequent speaker on AI in finance, representing the Oxford-Man Institute at academic conferences, industry panels and corporate events.

Afternoon networking break

15:30 - 16:00

Backtesting correlated quantities

17:00 - 17:30

Backtesting financial models over long horizons often leads to samples that are correlated. This poses a challenge as most standard statistical tests assume their independence. In the talk we will learn various strategies how to deal with this problem and understand the impact of these choices on the discriminatory power of the test.

  • arise naturally when backtesting over long horizons
  • review of statistical frameworks
  • discriminatory power analysis


Nikolai Nowaczyk

Technical specialist


Nikolai Nowaczyk is a Risk Management consultant who has advised more than 10 clients in over 20 projects around counterparty credit risk, xVA and model validation. He is broadly interested in classical methods of financial mathematics and statistics as well as data science and machine learning. Nikolai holds a PhD in mathematics from the University of Regensburg, has been an academic visitor to Imperial College London and likes to build bridges between academic research and practical applications.

Synthetic data

17:00 - 17:30

VolGAN: a generative model for arbitrage-free implied volatility

17:30 - 18:00

We introduce VolGAN, a generative model for arbitrage-free implied volatility surfaces. The model is trained on time series of implied volatility surfaces and is capable of generating dynamics scenarios for implied volatility surfaces with realistic dynamics. We illustrate the performance of the model by training it on SPX implied volatility time series and show that it can learn the covariance structure of co-movements in implied volatilities and generate realistic dynamics for the VIX volatility index.

Milena Vuletic

DPhil candidate, mathematics of random systems

Oxford University

Milena Vuletić is a DPhil Student in the Centre for Doctoral Training in Mathematics of Random Systems at the University of Oxford. She is supervised by Prof. Rama Cont and Prof. Mihai Cucuringu. Her research is focused on mathematical and data-driven modelling of multi-asset markets.


Quant Summit Europe cocktails
Network with peers over a drink

18:30 - 19:30