Preconference Day: Summits & Workshops - GMT (Greenwich Mean Time, GMTZ)
- Aous Labbane - Quant Investment Professional, Independent
Perspectives from across the industry. Aspects of risk and alpha.
- Erik Vynckier - Board Member, Chair of the Investment Committee, Foresters Friendly Society and Institute and Faculty of Actuaries
- Anton Merlushkin - Head of Quant Modelling & Analytics, Jain Global
- Lucette Yvernault - Head of Systematic Fixed Income, Fidelity International
- Aitor Muguruza Gonzalez - Chief Artificial Intelligence Officer, Kaiju Capital Management
How is the post-MiFID II era shaping up the landscape?
- Christopher Cormack - Independent, UK Centre for Greening Finance and Investment (CGFI)
This study offers an innovative approach to addressing the challenges of measuring workplace inclusion culture and provides concrete evidence of its impact on innovation, financial performance and stock returns. We leverage alternative data sources—such as online workforce profiles and employee reviews—and employ advanced data science and textual analysis techniques to construct meaningful inclusion culture signals.
- Andreas Theodoulou - Data Scientist, Citi
- Revant Nayar - Chief Investment Officer, Dilaton Technologies LLC Family Office
- Lucette Yvernault - Head of Systematic Fixed Income, Fidelity International
Perspectives on allocation strategy.
- Aous Labbane - Quant Investment Professional, Independent
- Ouaile El Fetouhi - Head of Portfolio Construction, Natixis Investment Managers Solutions (NIM Solutions)
We study how the global equity markets price physical climate risk associated with tropical cyclones. To assess firms' exposure to this risk, we use a bottom-up, forward-looking measure, based on probabilistic assessments of losses to firms’ geolocalized physical assets. The expected losses by all events or rare events are estimated considering various climate scenarios (RCP 2.6, 4.5, and 6.0). Additionally, we examine the impact of investors' concerns regarding cyclone risks on stock returns. We find that a one standard deviation higher exposure under RCP 4.5 is associated with a 1.5 to 2.2% higher annual return during periods of low cyclone concerns. However, during periods of heightened cyclone concerns, a one standard deviation higher in exposure leads to a drop of 0.8 to 1.7% annual return. Overall, our finding suggests that the global equity markets have started to incorporate physical climate risk related to tropical cyclones. However, during periods of increased cyclone activity, investors' concerns may cause a decrease in demand for stocks that are more exposed to this risk, resulting in a decline in their prices, even if these firms' facilities are not directly affected by cyclone events.
- Karen Huynh - Research Analyst, Amundi
- Jose Canals-Cerda - Senior Special Advisor, Supervision, Regulation and Credit, Federal Reserve Bank of Philadelphia
The “Interpretable Supervised Portfolio" framework is an innovative approach to asset allocation by leveraging the RuleFit algorithm to engineer optimal portfolio weights directly, prioritizing optimization before prediction. Contrary to traditional methods that often apply prediction before optimization, the proposed methodology ensures a more strategic asset allocation by using a supervised learning algorithm in a novel manner. The study employs an enhanced version of RuleFit, an intrinsically interpretable algorithm, to transform complex, non-linear predictive models into simpler, linear models that incorporate feature interactions from decision tree ensembles. This shift not only maintains statistical accuracy comparable to that of gradient boosting across various investment universes but also enhances transparency and trust in the models by offering direct insights into the relationship between inputs and outputs. A major finding is the linearized models' ability to perform on par with their non-linear counterparts, debunking the assumption that more complex models necessarily yield better predictions for optimal portfolio weights.
- Thomas Raffinot - Head of Quant Investment Signals, AXA Investment Managers
Many asset managers use sell-side brokers to buy and sell shares of common stock. Typically, orders are distributed between several brokers using so-called algo wheels. This allows asset managers to compare brokers, and, thereby, ensure they receive the best possible execution. This paper describes how reinforcement learning techniques can be used to optimally distribute orders between brokers. The methodologies find a balance between exploitation and exploration, i.e., the best performing broker receives most orders, and others receive sufficient orders to determine which broker performs best.
- Lars ter Braak - Trading Researcher, Robeco
- Martin van der Schans - Trading Researcher, Robeco
With increasing interest in equity portfolios that prioritize risk and volatility reduction, designing quantitative strategies with ultra-low beta (<0.1) is more crucial than ever. Generating attractive returns while managing downside risk effectively requires forward-thinking approaches, such as dynamic beta trading systems and targeted single-stock management. Quantmade, a global leader in wealthtech, has developed specialized expertise in these low-beta solutions.
- Michael Geke - CEO, Quantmade
- Aous Labbane - Quant Investment Professional, Independent
How has the use of AI in quantitative solutions developed in the past year? How are businesses using different solutions (LLMs vs GenAI) across the industry? What are the key standouts?
- Samar Gad - Associate Professor, Kingston Business School
- Federico Fontana - Chief Technology Officer, XAI Asset Management
- Ian McWilliam - Senior Researcher, Machine Learning, Aspect Capital
- Amal Moussa - Managing Director, Head of US Single Stocks Exotic Derivatives Trading, Goldman Sachs
Use cases, regulation and supervisory topics
- Matthias Fahrenwaldt - IT Policy and Supervision Suppport, Federal Financial Supervisory Authority (BaFin)
AI is changing the financial industry, and indeed, the world, at an ever accelerating pace. The current generation of AI models is different from the previous state of the art in important ways. Large transformer models are broadly capable – exhibiting state-of-the-art, and even human-level performance, on many tasks without problem-specific training – and uniquely accessible, allowing non-experts to interact with them using natural language.
Nonetheless, they also have important limitations. What is real, and what is hype? What are they capable of, and what can’t they do? In this talk, we will explore the current state of the art in AI applications, discuss the challenges of AI adoption in the enterprise, and talk about the risks and future directions of AI development.
- Gary Kazantsev - Head of Quant Technology Strategy, Bloomberg
This paper introduces a new quantitative toolkit for (reverse) stress testing of the Banking Book -- as required by EBA IRRBB regulations and the Basel Framework. Our toolkit combines classic yield curve modelling and valuation tools with classical Machine Learning clustering techniques to systematically identify blind spots in a bank’s balance sheet. I illustrate the model’s use by applying it to realistic balance sheets of two hypothetical banks and draw implications for risk management and policy making.
- Eric Schaanning - Group Head of Market and Valuation Risk Management, Nordea
- Jun Yuan - Managing Director, Global Risk Analytics, Royal Bank of Canada
- Jacky Chen - Managing Director, Total Portfolio Management, OPTrust
In this session, we'll dive into the transformative potential of Large Language Models (LLMs) in finance, moving beyond basic chatbot applications. The focus is on actionable workflows and strategies for leveraging LLMs to automate complex processes, enhance decision-making, and develop data-driven financial solutions. Attendees will gain insights into implementing LLMs to streamline operations, reduce manual effort, and boost efficiency in real-world financial settings.
- Fayssal El Mofatiche - Founder & CEO, Flowistic
- Nicole Königstein - Chief AI Officer, Head of AI & Quant Research, quantmate
- Stefano Pasquali - Head of Investment AI Modelling & Research team, BlackRock
- News data sourcing and processing.
- Prompt-based investment signals.
- LLM-based factor reduction to identify successful investment themes.
- Comparison to standard investment factors.
- Daniel Mayenberger - Head of Quants Risk as a Service Platform - Digital Products, J.P. Morgan
In the present work, we introduce and compare state-of-the-art algorithms, that are now classified under the name of machine learning, to price Asian and look-back products with early-termination features. These include randomized feed-forward neural networks, randomized recurrent neural networks, and a novel method based on signatures of the underlying price process. Additionally, we explore potential applications on callable certificates. Furthermore, we present an innovative approach for calculating sensitivities, specifically Delta and Gamma, leveraging Chebyshev interpolation techniques.
- Andrea Pallavicini - Head of Equity, FX and Commodity Models, Intesa Sanpaolo
Graph Neural Networks (GNNs) have recently shown significant promise in modeling complex relational data structures. In the realm of finance, where assets are often interconnected, capturing these relationships offers potential for improving time series prediction accuracy and explainability. This talk explores the application of GNNs to financial time series forecasting by representing assets and their interdependencies as dynamic graphs. We introduce novel architectures that integrate temporal dynamics with graph-based learning to model both the temporal evolution and the relational structure of financial data.
- Richard Turner - Managing Director, Currency Management, Mesirow Financial
- Blanka Horvath - Associate Professor in Mathematical and Computational Finance, University of Oxford
- Mikko Pakkanen - Reader in Data Science and Quantitative Finance, Imperial College
An overview of deep neural networks, activation functions, loss functions and training algorithms. Practical implementation in Pytorch.
- Blanka Horvath - Associate Professor in Mathematical and Computational Finance, University of Oxford
- Mikko Pakkanen - Reader in Data Science and Quantitative Finance, Imperial College
Unsupervised hedging and pricing of derivatives, high-frequency asset return prediction.
- Blanka Horvath - Associate Professor in Mathematical and Computational Finance, University of Oxford
- Mikko Pakkanen - Reader in Data Science and Quantitative Finance, Imperial College
Deep learning volatility and calibration of (rough) stochastic volatility models.
- Blanka Horvath - Associate Professor in Mathematical and Computational Finance, University of Oxford
- Mikko Pakkanen - Reader in Data Science and Quantitative Finance, Imperial College
Market generators and generative AI.
- Blanka Horvath - Associate Professor in Mathematical and Computational Finance, University of Oxford
- Mikko Pakkanen - Reader in Data Science and Quantitative Finance, Imperial College
- Blanka Horvath - Associate Professor in Mathematical and Computational Finance, University of Oxford
- Mikko Pakkanen - Reader in Data Science and Quantitative Finance, Imperial College
- Jim Gatheral - Presidential Professor of Mathematics, Baruch College, CUNY
- Shape of the volatility surface
- Scaling of implied volatility smiles
- Monofractal scaling of realized variance
- Estimation of H
- Realized variance forecasting
- Jim Gatheral - Presidential Professor of Mathematics, Baruch College, CUNY
- The forward variance curve
- Change of measure
- The rough Bergomi model
- The rough Heston model
- The quadratic rough Heston model
- Financial meaning of parameters
- Jim Gatheral - Presidential Professor of Mathematics, Baruch College, CUNY
- Affine forward intensity models
- Affine forward volatility models
- Diamonds and the exponentiation theorem
- The leverage swap
- Moment computations
- Jim Gatheral - Presidential Professor of Mathematics, Baruch College, CUNY
- Rational approximation of rough Heston
- The HQE scheme
- Parameter sensitivities
- Smile fitting
- Jim Gatheral - Presidential Professor of Mathematics, Baruch College, CUNY
- Jim Gatheral - Presidential Professor of Mathematics, Baruch College, CUNY
- Nicole Königstein - Chief AI Officer, Head of AI & Quant Research, quantmate