The brand-new Volatility and Risk in Quant Finance Conference takes place on 4th June 2025. Hear from industry leaders and expert practitioners as they explore the latest innovations and groundbreaking research in the field.
Video Type: Conference
AAD Applications as a Game Changer for Finance
Video Description
Portfolio Management Conference 2025
AAD (Algorithmic Adjoint Differentiation) is a powerful yet simple tool for AI and financial applications. This talk will explore the fundamentals of this technology and highlight its direct and indirect applications, which are truly groundbreaking.
Speaker Bio
Adil Reghai
Adil Reghai is a renowned expert in quantitative finance and artificial intelligence. With a strong academic background in mathematics and finance, he has held prominent roles in the financial industry, including positions at major banks and financial institutions. Adil is widely recognized for his contributions to the development and application of advanced mathematical techniques, such as Algorithmic Adjoint Differentiation (AAD), in solving complex financial problems. His work bridges the gap between cutting-edge technology and practical financial applications, making him a sought-after speaker and thought leader in the field. Adil is also an author and educator, sharing his expertise through publications, lectures, and workshops, inspiring the next generation of quants and AI practitioners.
Portfolio Management in Quant Finance Conference 2025
This is the highly anticipated return of the Portfolio Management in Quant Finance Conference, featuring an exciting lineup of groundbreaking discussions on the latest industry advancements.
Canonical Portfolios – Optimal Asset and Signal Combinations
Video Description
Portfolio Management Conference 2025
I will present a novel framework for analyzing the optimal asset and signal combination problem. The approach builds on the dynamic portfolio selection problem introduced by Brandt and Santa-Clara (2006) among others. We first reformulate their original investment problem into a tractable one that allows us to derive a closed-form expression for the optimal portfolio policy that is scalable to large cross-sectional financial applications. We then recast the problem of selecting a portfolio of correlated assets and signals into selecting a set of uncorrelated managed portfolios via Canonical Correlation Analysis (Hotelling, 1936). The new investment environment, where we consider the most forecastable portfolio and the second-most forecastable orthogonal portfolio, etc. This decomposition offers unique economic insights into the joint correlation structure of our optimal portfolio policy. We also operationalize our theoretical framework to bridge the gap between theory and practice, showcasing the improved performance of our proposed method over natural competing benchmarks.
Speaker Bio
Dr. Nick Firoozye
Dr. Nick Firoozye has over 20 years of experience in finance, in both buy and sell-side firms, including Lehman, Goldman, Deutsche Bank, Nomura, Sanford Bernstein, and Citadel, in research, structuring, and trading. He started finance in MBS/ABS, then EM Research, Credit, Asset Allocation, Rates Macro, RV Strategy and Trading, Distressed Debt Trading, and Vol Strategy, before moving into fully Systematic Trading in 2013. He is currently a Senior Researcher at Tradelink Worldwide Ltd, a proprietary trading firm. He is an Honorary Professor in Computer Science at University College London, focusing on Online Learning, RL, ML, and Statistics in Financial Trading. Nick has had 6 PhD students, four completed, with several working in Algo Trading roles. He co-authored a book, Managing Uncertainty, Mitigating Risk, about the role of uncertainty in finance after the Eurozone Crisis. Nick taught Algorithmic Trading Strategies, a PhD reading course in 2016 and has adapted it for an online course, creating an MSc course which he taught since 2018 to over 600 students. He is currently co-authoring a book on Algorithmic Trading. Nick got his PhD at Courant Institute, NYU, did postdocs at Univ Minn, Heriot-Watt, Bonn, NYU, and was an Asst Prof at University of Illinois, before leaving for Wall Street.
Quantitative Asset Allocation at a Swiss Insurer – Insights from Three Years
Video Description
A Talk from Portfolio Management Conference 2025
Three years ago, Helvetia Insurances’ Asset Management decided to build its own tool for Strategic Asset Allocation called Sally, a bespoke implementation of the model suggested by Black/Litterman (1992). During the three years that we have worked with Sally in a production environment, we have addressed quite a few use cases and encountered, and solved, several problems — for example, how to build risk factor models for illiquid asset classes, such as real estate. This talk discusses a few of those issues and our solutions, for example, how to define and measure key performance indicators, how to denoise the covariance matrix, or how we can use the implied equilibrium returns to find indicative expected returns for asset classes with sparse data.
Speaker Bio
Claus Huber
Claus is the Head of Quantitative Modelling & Analytics at Helvetia Insurances in Basel / Switzerland, where his team develops digital tools for Strategic and Tactical Asset Allocation, risk budgeting, overlay management, manager selection, visual representation of complex data structures, and a few more. In previous roles he developed new investment products for Quantitative Multi Asset Funds and, as Head of Digital Transformation, drove the development of new tools and data products that allow smart data usage and sharing, and advised clients on risk management and quantitative investment solutions. Claus has extensive experience as entrepreneur, risk manager, credit strategist, hedge fund analyst, and government bond trader and has worked for hedge funds, banks, and insurance companies.
Fast Times, Slow Times and Timescale Separation in Financial Timeseries Data
Video Description
A Talk from Portfolio Management Conference 2025
Quantitative strategies rely on stationarity of the underlying processes. Parameter drift eventually kills all strategies, how ever brilliant they may be. This talk will explore some methods for unmixing financial timeseries using intrinsic relaxation timescales, recovering any stationary processes, but also allowing us to separate “slow” from “fast” processes. This has various interesting implications for trading and portfolio allocation, which we will briefly discuss.
Speaker Bio
Dr. Jan Rosenzweig
Jan has been working in the financial markets for close to 20 years, as a quant, structurer, trader and portfolio manager. He worked at Credit Suisse, Rabobank, HSH Nordbank, IV Capital, Brancherose and Pine Tree. He has a PhD in applied maths from Cambridge and BSc in applied maths and computer science from Zagreb. Jan is based in London.
Jump Risk Premia in the Presence of Clustered Jumps
Video Description
A Talk from Portfolio Management Conference 2025
We introduce an option pricing model with clusters of jumps by incorporating a bivariate Hawkes process with exponential decay memory kernel. The Hawkes process characterises the self- and cross-excitement of positive jumps and negative jumps, allowing the model to effectively capture the volatile price dynamics, jump arrival times and implied volatility smiles often observed in cryptocurrencies such as Bitcoin (BTC). We derive jump risk premia for positive and negative jumps, defined as the discrepancies in jump measures between the objective measure and the risk-neutral measure. Our findings reveal that these jump risk premia: (i) provide insights on how the BTC options market reacted to major events, such as the COVID-19 outbreak and the FTX scandal; (ii) possess significant predictive power for delta-hedged option returns; and (iii) are indicators in explaining the volatile cost-of-carry implied from BTC futures prices.
Joint work with Francis Liu and Artur Sepp.
Speaker Bio
Professor Natalie Packham
Natalie Packham is Professor of Mathematics and Statistics at Berlin School of Economics and Law and Principal Researcher within the International Research Training Group “High Dimensional Nonstationary Time Series” (IRTG 1792) at Humboldt University Berlin. Natalie has several years of industry experience as a front office software engineer at an investment bank, and is frequently involved in industry-related research and consulting projects. Her research expertise includes Mathematical Finance, Financial Risk Management and Computational Finance, and her academic work has been published in Mathematical Finance, Finance & Stochastics, Quantitative Finance, Journal of Applied Probability and many other academic journals. She is associate editor of “Quantitative Finance”, “Methodology and Computing in Applied Probability” and “Digital Finance” and co-chair of the GARP Research Fellowship Advisory Board. Natalie holds an M.Sc. in Computer Science from the University of Bonn, a Master’s degree in Banking & Finance from Frankfurt School, and a Ph.D. in Quantitative Finance from Frankfurt School.
A Market Design to Trade Bundles of Securities and Minimal Exercise of American Options
Video Description
A Talk from Annual Conference 2024
Exchanges have a function of facilitating trades buy allowing for buyers and sellers of a security to meet and trade. However many strategies require the trade of not only one security but simultaneous execution of several legs, involving several securities. Examples include pair trading, equity portfolios, multi–maturities Futures strategies and options combinations. We call these multi-legs trades bundles. Not executing the various legs at the same time creates a risk of adverse price movement before the full completion. It can be eliminated by crossing the spreads and posting market orders but it incurs cost. We present a mechanism that allows for market participants to post one-sided orders on arbitrary bundles and then the matching engine computes (case of buy order) the cheapest super-replication of the posted bundle by a portfolio of other already posted bundles. So the features of the algorithm is that 1) it does not match security by security but rather a bundle with a collection of other bundles and 2) it can be a “super-match” in the sense that the super-replication may give the additional benefit of a positive residual. We detail the algorithm and apply this methodology to show how it can in certain cases improve the price of option combinations. This approach has an economic value as it allows for more trades to occur at mutually desirable prices.
We focus in particular on bundles of options of a same maturity on a stock. The dominance in terms of profile is easy to understand in the case of European options and we show how it can be generalized to American options.
Speaker Bio
Bruno Dupire
Paul is the founder of the Certificate in Quantitative Finance and Wilmott.com and he is internationally renowned as a leading expert on quantitative finance. His research work is extensive, with more than 100 articles in leading mathematical and finance journals, as well as several internationally acclaimed books on mathematical modeling and derivatives, including the best-selling Paul Wilmott On Quantitative Finance, published by John Wiley & Sons.
Machine Learning for Factor-based Commodities Investing
Video Description
A Talk from Annual Conference 2024
In the past decade, machine learning techniques have been widely adopted in the equities space, driving innovation both in academia and among practitioners. Leveraging the rich data environment of single-name equities, these advancements have unlocked new insights in quantitative finance. However, the application of machine learning in commodities futures markets remains underexplored, primarily due to the “data-poor” nature of this asset class compared to equities. This research investigates the use of machine learning in systematic commodities futures trading, expanding beyond the traditional price-based strategies that have dominated the field. We integrate well-established factor-based signals, commonly used in commodities investing, as inputs into our machine learning models. Our findings demonstrate that machine learning enhances the evolution of systematic commodities strategies, offering a more adaptable and comprehensive framework for forecasting market behavior and constructing long-short portfolios. Utilizing a 30-year dataset of 44 commodities futures contracts, we train machine learning models to predict future performance and build long-short portfolios. The results indicate that these models not only rival the performance of traditional systematic approaches but also provide deeper insights into market dynamics and critical factors influencing commodities asset pricing.
Speaker Bio
Tony Guida
Tony Guida is a Quantitative Portfolio Manager and researcher. He began his career at Unigestion in 2006, joining the quantitative equity team as a research analyst. He later became a member of the Research and Investment Committee for Minimum Variance Strategies, where he led the factor investing research group for institutional clients. In 2015, Tony moved to Edhec Risk Scientific Beta as a Senior Consultant, specializing in risk allocation and factor strategies. In 2016, he joined a major UK pension fund to build an in-house systematic equity strategy, co-managing £8 billion as a Senior Quantitative Portfolio Manager.
In January 2019, Tony joined RAM Active Investments as a Senior Quantitative Researcher in equities, eventually co-heading the systematic macro hedge fund offering. In 2023, he co-founded a hedge fund, serving as Co-Head of Research and Chief Data Scientist, focusing on systematic cross-asset strategies.
Tony holds Bachelor’s and Master’s degrees in Econometrics, Economics, and Finance from the University of Savoy, France. He has co-authored and edited several books on the application of machine learning in finance, including:
• Big Data and Machine Learning in Quantitative Investment (Wiley, 2018)
• Machine Learning for Factor Investing, R Version (CRC, 2020)
• Machine Learning for Factor Investing, Python Version (CRC, 2023)
He is also an advisory board member for the Financial Data Professional Institute and serves as a reviewer for machine learning papers in several academic journals.
Annual Quant Insights Conference 2024
Here’s the Annual Quant Insights Conference 2024 with a brand new line-up of talks from leading quant finance experts.