arXiv subjects

Damien Challet

Publications and source records attributed to Damien Challet.

At least 19 recordsLinked to original sources

arXiv2026

Optimal execution on Uniswap v2/v3 under transient price impact

We study the optimal liquidation of a large position on Uniswap v2 and Uniswap v3 in discrete time. The instantaneous price impact is derived from the AMM pricing rule. Transient impact is modeled to capture either exponential or approximately power-law decay, together with a permanent component. In the Uniswap v2 setting, we obtain optimal strategies in closed-form under general price dynamics. For Uniswap v3, we consider a two-layer liquidity framework, which naturally extends to multiple layers. We address the problem using dynamic programming under geometric Brownian motion dynamics and approximate the solution numerically using a discretization scheme. We obtain optimal strategies akin to classical ones in the LOB literature, with features specific to Uniswap. In particular, we show how the liquidity profile influences them.

q-fin.MF↗

arXiv2025

Noise-proofing Universal Portfolio Shrinkage

We enhance the Universal Portfolio Shrinkage Approximator (UPSA) of Kelly et al. (2023) by making it more robust with respect to estimation noise and covariate shift. UPSA optimizes the realized Sharpe ratio using a relatively small calibration window, leveraging ridge penalties and cross-validation to yield better portfolios. Yet, it still suffers from the staggering amount of noise in financial data. We propose two methods to make UPSA more robust and improve its efficiency: time-averaging of the optimal penalty weights and using the Average Oracle correlation eigenvalues to make covariance matrices less noisy and more robust to covariate shift. Combining these two long-term averages outperforms UPSA by a large margin in most specifications.

q-fin.RM↗

arXiv2025

Learning the Spoofability of Limit Order Books With Interpretable Probabilistic Neural Networks

This paper investigates real-time detection of spoofing activity in limit order books, focusing on cryptocurrency centralized exchanges. We first introduce novel order flow variables based on multi-scale Hawkes processes that account both for the size and placement distance from current best prices of new limit orders. Using a Level-3 data set, we train a neural network model to predict the conditional probability distribution of mid price movements based on these features. Our empirical analysis highlights the critical role of the posting distance of limit orders in the price formation process, showing that spoofing detection models that do not take the posting distance into account are inadequate to describe the data. Next, we propose a spoofing detection framework based on the probabilistic market manipulation gain of a spoofing agent and use the previously trained neural network to compute the expected gain. Running this algorithm on all submitted limit orders in the period 2024-12-04 to 2024-12-07, we find that 31% of large orders could spoof the market. Because of its simple neuronal architecture, our model can be run in real time. This work contributes to enhancing market integrity by providing a robust tool for monitoring and mitigating spoofing in both cryptocurrency exchanges and traditional financial markets.

q-fin.TR↗

arXiv2025

Optimal risk-aware interest rates for decentralized lending protocols

Interest rates in decentralized lending protocols are set algorithmically and adjust to supply and demand for liquidity. In this study, we propose an optimal interest rate model that maximizes the expected lender wealth while incorporating penalties for liquidity risk and interest rate stabilization. This objective benefits both sides of the market: it improves yield and reduces liquidity risk for lenders, while encouraging borrower activity indirectly through higher utilization and directly through stabilized borrowing costs. The dynamics of the utilization rate are modeled using point processes whose intensities depend on the interest rate. When intensities are linear, the optimal interest rate model is derived from a system of Riccati-type ODEs. In the nonlinear case, we approximate it using a Monte-Carlo estimator coupled with deep learning techniques. Finally, using block-by-block data, we conduct a risk-adjusted profit and loss analysis to compare industry-standard interest rate models to the deep learning-based one.

q-fin.MF↗

arXiv2024

A minimal model of money creation under regulatory constraints

We propose a minimal model of the secured interbank network able to shed light on recent money markets puzzles. We find that excess liquidity emerges due to the interactions between the reserves and liquidity ratio constraints; the appearance of evergreen repurchase agreements and collateral re-use emerges as a simple answer to banks' counterparty risk and liquidity ratio regulation. In line with prevailing theories, re-use increases with collateral scarcity. In our agent-based model, banks create money endogenously to meet the funding requests of economic agents. The latter generate payment shocks to the banking system by reallocating their deposits. Banks absorbs these shocks thanks to repurchase agreements, while respecting reserves, liquidity, and leverage constraints. The resulting network is denser and more robust to stress scenarios than an unsecured one; in addition, the stable bank trading relationships network exhibits a core-periphery structure. Finally, we show how this model can be used as a tool for stress testing and monetary policy design.

econ.GN↗

arXiv2024

Stylized facts in money markets: an empirical analysis of the eurozone data

Using the secured transactions recorded within the Money Markets Statistical Reporting database of the European Central Bank, we test several stylized facts regarding interbank market of the 47 largest banks in the eurozone. We observe that the surge in the volume of traded evergreen repurchase agreements followed the introduction of the LCR regulation and we measure a rate of collateral re-use consistent with the literature. Regarding the topology of the interbank network, we confirm the high level of network stability but observe a higher density and a higher in- and out-degree symmetry than what is reported for unsecured markets.

econ.GN↗

arXiv2024

Consistent time travel for realistic interactions with historical data: reinforcement learning for market making

Reinforcement learning works best when the impact of the agent's actions on its environment can be perfectly simulated or fully appraised from available data. Some systems are however both hard to simulate and very sensitive to small perturbations. An additional difficulty arises when a RL agent is trained offline to be part of a multi-agent system using only anonymous data, which makes it impossible to infer the state of each agent, thus to use data directly. Typical examples are competitive systems without agent-resolved data such as financial markets. We introduce consistent data time travel for offline RL as a remedy for these problems: instead of using historical data in a sequential way, we argue that one needs to perform time travel in historical data, i.e., to adjust the time index so that both the past state and the influence of the RL agent's action on the system coincide with real data. This both alleviates the need to resort to imperfect models and consistently accounts for both the immediate and long-term reactions of the system when using anonymous historical data. We apply this idea to market making in limit order books, a notoriously difficult task for RL; it turns out that the gain of the agent is significantly higher with data time travel than with naive sequential data, which suggests that the difficulty of this task for RL may have been overestimated.

q-fin.TR↗

arXiv2024

When is cross impact relevant?

Trading pressure from one asset can move the price of another, a phenomenon referred to as cross impact. Using tick-by-tick data spanning 5 years for 500 assets listed in the United States, we identify the features that make cross-impact relevant to explain the variance of price returns. We show that price formation occurs endogenously within highly liquid assets. Then, trades in these assets influence the prices of less liquid correlated products, with an impact velocity constrained by their minimum trading frequency. We investigate the implications of such a multidimensional price formation mechanism on interest rate markets. We find that the 10-year bond future serves as the primary liquidity reservoir, influencing the prices of cash bonds and futures contracts within the interest rate curve. Such behaviour challenges the validity of the theory in Financial Economics that regards long-term rates as agents anticipations of future short term rates.

q-fin.TR↗

arXiv2024

Equity auction dynamics: latent liquidity models with activity acceleration

Equity auctions display several distinctive characteristics in contrast to continuous trading. As the auction time approaches, the rate of events accelerates causing a substantial liquidity buildup around the indicative price. This, in turn, results in a reduced price impact and decreased volatility of the indicative price. In this study, we adapt the latent/revealed order book framework to the specifics of equity auctions. We provide precise measurements of the model parameters, including order submissions, cancellations, and diffusion rates. Our setup allows us to describe the full dynamics of the average order book during closing auctions in Euronext Paris. These findings support the relevance of the latent liquidity framework in describing limit order book dynamics. Lastly, we analyze the factors contributing to a sub-diffusive indicative price and demonstrate the absence of indicative price predictability.

q-fin.TR↗

arXiv2024

Can ChatGPT Compute Trustworthy Sentiment Scores from Bloomberg Market Wraps?

We used a dataset of daily Bloomberg Financial Market Summaries from 2010 to 2023, reposted on large financial media, to determine how global news headlines may affect stock market movements using ChatGPT and a two-stage prompt approach. We document a statistically significant positive correlation between the sentiment score and future equity market returns over short to medium term, which reverts to a negative correlation over longer horizons. Validation of this correlation pattern across multiple equity markets indicates its robustness across equity regions and resilience to non-linearity, evidenced by comparison of Pearson and Spearman correlations. Finally, we provide an estimate of the optimal horizon that strikes a balance between reactivity to new information and correlation.

q-fin.ST↗

arXiv2023

Covariance matrix filtering and portfolio optimisation: the Average Oracle vs Non-Linear Shrinkage and all the variants of DCC-NLS

The Average Oracle, a simple and very fast covariance filtering method, is shown to yield superior Sharpe ratios than the current state-of-the-art (and complex) methods, Dynamic Conditional Covariance coupled to Non-Linear Shrinkage (DCC+NLS). We pit all the known variants of DCC+NLS (quadratic shrinkage, gross-leverage or turnover limitations, and factor-augmented NLS) against the Average Oracle in large-scale randomized experiments. We find generically that while some variants of DCC+NLS sometimes yield the lowest average realized volatility, albeit with a small improvement, their excessive gross leverage and investment concentration, and their 10-time larger turnover contribute to smaller average portfolio returns, which mechanically result in smaller realized Sharpe ratios than the Average Oracle. We also provide simple analytical arguments about the origin of the advantage of the Average Oracle over NLS in a changing world.

q-fin.ST↗

arXiv2023

Price impact in equity auctions: zero, then linear

Using high-quality data, we report several statistical regularities of equity auctions in the Paris stock exchange. First, the average order book density is linear around the auction price at the time of auction clearing and has a large peak at the auction price. While the peak is due to slow traders, the order density shape is the result of subtle dynamics. The impact of a new market order or cancellation at the auction time can be decomposed into three parts as a function of the size of the additional order: (1) zero impact, caused by the discrete nature of prices, sometimes up to a surprisingly large additional volume relative to the auction volume (2) linear impact for additional orders up to a large fraction of the auction volume (3) for even larger orders price impact is non-linear, frequently super-linear.

q-fin.ST↗

arXiv2023

Recurrent Neural Networks with more flexible memory: better predictions than rough volatility

We extend recurrent neural networks to include several flexible timescales for each dimension of their output, which mechanically improves their abilities to account for processes with long memory or with highly disparate time scales. We compare the ability of vanilla and extended long short term memory networks (LSTMs) to predict asset price volatility, known to have a long memory. Generally, the number of epochs needed to train extended LSTMs is divided by two, while the variation of validation and test losses among models with the same hyperparameters is much smaller. We also show that the model with the smallest validation loss systemically outperforms rough volatility predictions by about 20% when trained and tested on a dataset with multiple time series.

q-fin.ST↗

arXiv2023

Dissecting the explanatory power of ESG features on equity returns by sector, capitalization, and year with interpretable machine learning

We systematically investigate the links between price returns and Environment, Social and Governance (ESG) scores in the European equity market. Using interpretable machine learning, we examine whether ESG scores can explain the part of price returns not accounted for by classic equity factors, especially the market one. We propose a cross-validation scheme with random company-wise validation to mitigate the relative initial lack of quantity and quality of ESG data, which allows us to use most of the latest and best data to both train and validate our models. Gradient boosting models successfully explain the part of annual price returns not accounted for by the market factor. We check with benchmark features that ESG data explain significantly better price returns than basic fundamental features alone. The most relevant ESG score encodes controversies. Finally, we find the opposite effects of better ESG scores on the price returns of small and large capitalization companies: better ESG scores are generally associated with larger price returns for the latter and reversely for the former.

q-fin.PM↗

arXiv2023

Reactive Global Minimum Variance Portfolios with $k-$BAHC covariance cleaning

We introduce a $k$-fold boosted version of our Boostrapped Average Hierarchical Clustering cleaning procedure for correlation and covariance matrices. We then apply this method to global minimum variance portfolios for various values of $k$ and compare their performance with other state-of-the-art methods. Generally, we find that our method yields better Sharpe ratios after transaction costs than competing filtering methods, despite requiring a larger turnover.

q-fin.PM↗

arXiv2023

Cleaning the covariance matrix of strongly nonstationary systems with time-independent eigenvalues

We propose a data-driven way to reduce the noise of covariance matrices of nonstationary systems. In the case of stationary systems, asymptotic approaches were proved to converge to the optimal solutions. Such methods produce eigenvalues that are highly dependent on the inputs, as common sense would suggest. Our approach proposes instead to use a set of eigenvalues totally independent from the inputs and that encode the long-term averaging of the influence of the future on present eigenvalues. Such an influence can be the predominant factor in nonstationary systems. Using real and synthetic data, we show that our data-driven method outperforms optimal methods designed for stationary systems for the filtering of both covariance matrix and its inverse, as illustrated by financial portfolio variance minimization, which makes out method generically relevant to many problems of multivariate inference.

stat.AP↗

arXiv2022

Non-linear shrinkage of the price return covariance matrix is far from optimal for portfolio optimisation

Portfolio optimization requires sophisticated covariance estimators that are able to filter out estimation noise. Non-linear shrinkage is a popular estimator based on how the Oracle eigenvalues can be computed using only data from the calibration window. Contrary to common belief, NLS is not optimal for portfolio optimization because it does not minimize the right cost function when the asset dependence structure is non-stationary. We instead derive the optimal target. Using historical data, we quantify by how much NLS can be improved. Our findings reopen the question of how to build the covariance matrix estimator for portfolio optimization in realistic conditions.

q-fin.PM↗

arXiv2022

Statistical inference of lead-lag at various timescales between asynchronous time series from p-values of transfer entropy

Symbolic transfer entropy is a powerful non-parametric tool to detect lead-lag between time series. Because a closed expression of the distribution of Transfer Entropy is not known for finite-size samples, statistical testing is often performed with bootstraps whose slowness prevents the inference of large lead-lag networks between long time series. On the other hand, the asymptotic distribution of Transfer Entropy between two time series is known. In this work, we derive the asymptotic distribution of the test for one time series having a larger Transfer Entropy than another one on a target time series. We then measure the convergence speed of both tests in the small sample size limits via benchmarks. We then introduce Transfer Entropy between time-shifted time series, which allows to measure the timescale at which information transfer is maximal and vanishes. We finally apply these methods to tick-by-tick price changes of several hundreds of stocks, yielding non-trivial statistically validated networks.

q-fin.ST↗