Albert Menkveld (Vrije Universiteit Amsterdam): Does News Trading Crowd Out Value Trading? Evidence from a Structural Estimation.
Technology produces data streams at lower cost, which intensifies "news" trading. Does this trading on value changes crowd out trading on value levels, i.e., value trading? And, what is the net effect on liquidity and market efficiency? We propose a Kyle (1985)-type model to study these questions, both theoretically and empirically. The latter is possible, because closed-form solutions admit structural estimation. Implementing it on a 2013-2023 proprietary sample covering trades in the EURO STOXX 50 Index Futures reveals how liquidity, market efficiency, and the profit of both types of informed trading changed over time, and why.
Joint work with Ion Lucas Saru.
Anastasia Borovykh (CFM): Building a Continuously Learning Financial Analyst
When a firm hires a financial research analyst, the analyst is expected to deepen their knowledge of particular markets and asset classes, develop and test fundamental trade ideas, observe how those ideas play out, and become more capable through accumulated experience. Current AI systems acquire most of their capabilities during pre- and post-training. To deploy them effectively in the real world, we need systems that can continually learn: increasing their competence as the world evolves. We begin by reviewing prompt- and weight-based continual learning methods and stress-test them under different forms of environmental change: domain shift, temporal drift, and action-induced state change. We then apply these methods to the construction of a continually learning financial analyst. We study which approaches remain effective under the noisy, time-varying feedback, where current methods fall short, and under what conditions such an analyst can outperform human experts.
Jean-Philippe Bouchaud (CFM): The Subtle Interplay Between Square-Root Impact, Order Imbalance & Volatility
We will discuss an attempt to reconcile several apparently contradictory observations in market microstructure: is the famous "square-root law" of metaorder impact, which decays with time, compatible with the random-walk nature of prices and the linear impact of order imbalances? Can one entirely explain the volatility of prices as resulting from the flow of uninformed metaorders that mechanically impact them? We introduce a new theoretical framework to describe metaorders with different signs, sizes and durations, which all impact prices as a square-root of volume but with a subsequent time decay. We show that, as in the original propagator model, price diffusion is ensured by the long memory of cross-correlations between metaorders. In order to account for the effect of strongly fluctuating volumes q of individual trades, we further introduce two q-dependent exponents, which allow us to describe how the moments of generalized volume imbalance and the correlation between price changes and generalized order flow imbalance scale with T. We predict in particular that the corresponding power-laws depend in a non-monotonic fashion on a parameter a, which allows one to put the same weight on all child orders or to overweight large ones, a behavior that is clearly borne out by empirical data.We also predict that the correlation between price changes and volume imbalances should display a maximum as a function of a, which again matches observations. We argue that our results support the "Order-Driven" theory of excess volatility, and are at odds with the idea that a "fundamental" component accounts for a large share of the volatility of financial markets.
Joint work with Guillaume Maitrier.
Jonathan Hall (Bank of England): Monsters in the Deep Revisited
This talk will be a follow-up to the original one here
Marcel Nutz (Columbia University): Optimal Fees for Liquidity Provision in Automated Market Makers
Passive liquidity providers (LPs) in automated market makers (AMMs) face losses due to adverse selection (LVR), which static trading fees often fail to offset in practice. We study the key determinants of LP profitability in a dynamic reduced-form model where an AMM operates in parallel with a centralized exchange (CEX), traders route their orders optimally to the venue offering the better price, and arbitrageurs exploit price discrepancies. Using large-scale simulations and real market data, we analyze how LP profits vary with market conditions such as volatility and trading volume, and characterize the optimal AMM fee as a function of these conditions. We highlight the mechanisms driving these relationships through extensive comparative statics, and confirm the model's relevance through market data calibration. A key trade-off emerges: fees must be low enough to attract volume, yet high enough to earn sufficient revenues and mitigate arbitrage losses. We find that under normal market conditions, the optimal AMM fee is competitive with the trading cost on the CEX and remarkably stable, whereas in periods of very high volatility, a high fee protects passive LPs from severe losses. These findings suggest that a threshold-type dynamic fee schedule is both robust enough to market conditions and improves LP outcomes.
Joint work with Steven Campbell, Philippe Bergault and Jason Milionis.
Marcin Kacperczyk (Imperial): Adaptive Markets Hypothesis and Conditional Ecology Performance: Evidence from Machine-Learning Market-Timing Strategies
We evaluate traditional and modern machine-learning market-timing strategies from 1970 to 2024 under a common information set, portfolio rule, and realistic transaction costs. Unconditionally, none of the strategies delivers robust abnormal performance relative to buy-and-hold, consistent with the Efficient Markets Hypothesis (EMH). Conditioning on an ex-ante ecology vector capturing crowding, funding, liquidity, and volatility, performance, risk, and cross-strategy co-movement vary systematically, with different model families specializing in distinct habitats. These state-dependent patterns are difficult to reconcile with a purely risk-based EMH but align closely with the Adaptive Markets Hypothesis (AMH).
Joint wort with Franklin Allen and K S Sesh Kumar.
Mathieu Rosenbaum (Université Paris Dauphine): A Unified Theory of Order Flow, Market Impact and Volatility
We propose a microstructural model for the order flow in financial markets that distinguishes between core orders and reaction flow, both modeled as Hawkes processes. This model has a natural scaling limit that reconciles a number of salient empirical properties: persistent signed order flow, rough trading volume and volatility, and power-law market impact. In our framework, all these quantities are pinned down by a single statistic H which measures the persistence of the core flow. Specifically, the signed flow converges to the sum of a fractional process with Hurst index H and a martingale, while the limiting traded volume is a rough process with Hurst index H-1/2. No-arbitrage constraints imply that volatility is rough, with Hurst parameter 2H-3/2, and that the price impact of trades follows a power law with exponent 2-2H.The analysis of signed order flow data yields an estimate H approx 3/4. This is not only consistent with the square-root law of market impact, but also turns out to match estimates for the roughness of traded volumes and volatilities remarkably well.
Joint work with Johannes Muhle-Karbe, Youssef Ouazzani Chahdi, and Gregoire Szymanski.
Pierre Collin-Dufresne (EPF Lausanne): Asymmetric Reversals
Short-term return reversal is one of the most robust asset-pricing anomalies, and is commonly linked to liquidity provision. We decompose individual firm stock returns into two distinct components: SYS, the component of returns that can be linked to public information releases; and an orthogonal residual RES. The RES component reverses, while the SYS component exhibits continuation. Moreover, the residual reversals are highly asymmetric: positive residual shocks reverse much more slowly than negative shocks. A return factor based on asymmetric idiosyncratic reversal (AIR) is a powerful short-term predictor of future returns, and the factor subsumes the idiosyncratic volatility factor and a broad set of other short-horizon anomalies-based factors. Our findings suggest that asymmetric idiosyncratic reversal is the primary driver of short-term return predictability.
Joint work with Federico Baldi-Lanfranchi and Kent Daniel.
Renyuan Xu (Stanford): Generative Diffusion Models in Finance
Generative AI has the potential to transform how we model, understand, and simulate complex financial systems. Yet financial markets pose fundamental challenges: they are nonstationary, exhibit strong temporal and cross-sectional dependence, undergo regime changes and rare events, and are governed by economic mechanisms and market constraints. In this talk, I will present a principled framework for diffusion-based generative modeling that reflects these distinctive features. I will discuss how low-dimensional factor structures can make the generation of high-dimensional asset returns statistically tractable; how sequential models can respect the flow of information and the non-anticipative structure of financial time series; and how tools from stochastic analysis can steer generation toward rare events or constrained scenarios. Together, these directions seek to establish a scientific foundation for generative modeling of financial data by integrating statistical structure, temporal dynamics, and economic constraints.
Joint work with Minshuo Chen, Yumin Xu and Ruixun Zhang.
Svetlana Bryzgalova (London Business School): Macro Strikes Back: Term Structure of Risk Premia
We provide a novel priced Wold representation that, using the pricing restrictions of a large cross-section of asset returns, sharply identifies shocks common to financial markets and the macroeconomy, and their propagation. These shocks slowly propagate through major macro aggregates, account for 20-47% of their variation and most of their predictability, and trace their business cycle, disciplining all equilibrium models. This propagation, not the overall persistence of macro quantities, yields short-run macro risk premia that are negligible, yet match the equity premium at business cycle horizons. Validating the method, the model-implied prices of dividend strips match observed out-of-sample forward equity yields and their term structure, both conditionally and unconditionally. By identification through elimination, we rule out productivity, investment-specific technology, and pure preference shocks as likely origins of the business cycle. The data point to demand/belief shocks and cost-push shocks propagating through real, nominal, and informational rigidities.
Joint work with Jiantao Huang and Christian Julliard.
Thomas Bromley (Quantum Motion): An Introduction to Quantum Computing and its Applications in Finance
Quantum computers hold the potential to accelerate solving problems across science and industry, but how can they make an impact specifically in finance? This talk begins with an overview of the quantum computing industry, from promising hardware modalities to challenges in quantum error correction. We then survey the potential use cases of a quantum computer in finance, using derivates pricing via quantum-enhanced Monte Carlo estimation as a case study to understand potential speedups, drawbacks, and algorithm design tradeoffs.
Zoltan Eisler (Imperial): Is Trend Still Your Friend? A Microstrucrual Account of the Demise of Short-Term Trend Following
Systematic trend following has, on average, been profitable for at least two centuries; yet since approximately 2009, short-term trends have ceased to deliver reliable returns. Using an industry-representative CTA proxy over the period 1995-2025, we document the break and evaluate four candidate explanations - capacity constraints, market electronification, a regime change in CTA-versus-order-flow interactions, and a microstructural mechanism. Our central empirical finding is that the cross-sectional variable distinguishing degraded from surviving trends is the volatility-normalised tick size: Post-2008 trend PnL has collapsed on small-tick contracts across all signal horizons, while remaining essentially intact on large-tick ones. We interpret this result through a self-fulfilling feedback loop: trend signals trigger directional trades, whose market impact reinforces the very price moves that generated the signal. We argue that post-crisis, HFT-dominated market markers withdraw liquidity in front of predictable directional flow. Following this change, on large-tick contracts, residual depth remains sufficient, and the loop continues to operate, while it is broken on small-tick contracts.
Joint work with Jutta Kurth, Adam Rej and Jean-Philippe Bouchaud.