arXiv subjects

Mihai Cucuringu

Publications and source records attributed to Mihai Cucuringu.

At least 19 recordsLinked to original sources

arXiv2026

Spectral clustering of network time series via the sample covariance matrix

Spectral clustering for community detection is analysed in multivariate time series models whose dependence structure is determined by an unobserved stochastic blockmodel. We establish that spectral clustering of the sample covariance matrix achieves exact recovery of the underlying communities. The recovery rates depend explicitly on the network size, sample length, block separation, and degree of data dependence. This demonstrates that exact community recovery under a stochastic blockmodel is possible even when the adjacency matrix is unobserved. Our theory provides extensions of both classical and fine-grained matrix perturbation theory to the setting of dependent data, which may be of independent interest.

math.ST↗

arXiv2026

Gaussian Boson Sampling for Asset Clustering in Statistical Arbitrage Portfolios

Gaussian Boson Sampling (GBS) provides a native photonic quantum heuristic for sampling dense subgraphs from adjacency matrices, offering a scalable physical approach to combinatorial graph search problems. Simultaneously, correlation matrix clustering algorithms, such as Spectral and SPONGE, have established robust benchmarks for identifying co-moving assets from correlation matrices in statistical arbitrage (StatArb) strategies. In this work, we map S&P 500 residual correlation data into GBS-compatible adjacency matrices. We benchmark those classical clustering algorithms against two quantum clustering algorithms, GBS Boost and our novel GBS Roots, to construct dynamic, market-neutral portfolios over a rolling one-year window. Simulations across distinct macroeconomic regimes reveal that quantum clustering generates superior alpha within large stock universes during periods of high volatility, effectively isolating structural market idiosyncrasies. Crucially, this economic advantage persists under simulated low-loss conditions and extends into high-loss regimes via the application of coherent displacement to compensate for photon loss. Our findings underscore the efficacy of GBS-derived graph clustering in constructing robust StatArb portfolios, establishing a quantum foundation for broader quantitative finance applications.

quant-ph↗

arXiv2026

Temporal Motif Signatures for Temporal Graph Neural Networks

Real temporal interaction streams carry predictive structure in short-horizon motif patterns -- repetition, reciprocity, star diversity, triadic flow -- that vanilla temporal graph neural networks (TGNNs) often fail to expose to their edge scorers. We show this concretely on MOOC interaction prediction, where a small four-feature family of past-window star counts already delivers most of the lift over a strong static GNN. Across a wide set of real and synthetic temporal datasets we find that motif activity organizes consistently along three scale-stable axes (dyadic recency/reciprocity, star diversity, triadic flow), and we use this empirical structure to design a compact 13-coordinate, leakage-safe, candidate-local motif feature map h(u, v, t) that linearly embeds into any static or temporal encoder without architectural changes. A temporal Weisfeiler-Leman (WL) analysis places the augmentation relative to the first level of an anchored temporal-WL hierarchy and exhibits a candidate-anchored pair on which motif features distinguish. We demonstrate empirically that the same augmentation consistently lifts performance across heterogeneous tasks: TGB link-property prediction across all five baselines, edge classification on Bitcoin Alpha/OTC and MOOC, and graph-level classification of synthetic temporal generators.

cs.LG↗

arXiv2026

Macro-aware time series forecasting via hierarchical mixed-frequency attention models

Deep learning models show promise in financial forecasting, yet their generalization is often undermined by small datasets, noisy signals, and non-stationarity. While meta-learning and related techniques mitigate some of these issues, they typically do not account for a core limitation in macro-financial prediction: the scarcity of distinct macroeconomic regimes that drive asset returns. We introduce HANET (Hierarchical Attention Network), a hybrid LSTM-based architecture that integrates macroeconomic domain knowledge through attention over long-run macro contexts while preserving high-frequency market dynamics. HANET organizes information in a hierarchical mixed-frequency structure, with daily asset-return signals nested within monthly macroeconomic windows, and introduces a Hierarchical Cross-Attention mechanism that reconciles low-frequency macro signals with high-frequency returns without discarding granular daily information. By framing regime selection as attention over macroeconomic contexts, the model adapts to scarce and shifting regimes. Empirically, across 55 liquid futures spanning multiple asset classes, HANET consistently outperforms neural forecasters that ignore macroeconomic information, particularly during turbulent periods, improving risk-adjusted returns and mitigating losses. Ablation studies show that these gains rely on structured macro conditioning rather than naive feature augmentation: an LSTM with the same macro representation performs poorly, and shuffling macro contexts substantially degrades performance. Finally, HANET provides interpretability through attention weights, highlighting which historical regimes are most influential for each forecast and linking macro conditions to portfolio outcomes. These results establish HANET as a systematic approach to integrating macroeconomic information into attention-based deep learning for financial forecasting.

q-fin.ST↗

arXiv2026

A Bipartite Graph Approach to U.S.-China Cross-Market Return Forecasting

This paper studies cross-market return predictability through a machine learning framework that preserves economic structure. Exploiting the non-overlapping trading hours of the U.S. and Chinese equity markets, we construct a directed bipartite graph that captures time-ordered predictive linkages between stocks across markets. Edges are selected via rolling-window hypothesis testing, and the resulting graph serves as a sparse, economically interpretable feature-selection layer for downstream machine learning models. We apply a range of regularized and ensemble methods to forecast open-to-close returns using lagged foreign-market information. Our results reveal a pronounced directional asymmetry: U.S. previous-close-to-close returns contain substantial predictive information for Chinese intraday returns, whereas the reverse effect is limited. This informational asymmetry translates into economically meaningful performance differences and highlights how structured machine learning frameworks can uncover cross-market dependencies while maintaining interpretability.

cs.LG↗

arXiv2026

Data-Driven Graph Filters via Adaptive Spectral Shaping

We introduce Adaptive Spectral Shaping, a data-driven framework for graph filtering that learns a reusable baseline spectral kernel and modulates it with a small set of Gaussian factors. The resulting multi-peak, multi-scale responses allocate energy to heterogeneous regions of the Laplacian spectrum while remaining interpretable via explicit centers and bandwidths. To scale, we implement filters with Chebyshev polynomial expansions, avoiding eigendecompositions. We further propose Transferable Adaptive Spectral Shaping (TASS): the baseline kernel is learned on source graphs and, on a target graph, kept fixed while only the shaping parameters are adapted, enabling few-shot transfer under matched compute. Across controlled synthetic benchmarks spanning graph families and signal regimes, Adaptive Spectral Shaping reduces reconstruction error relative to fixed-prototype wavelets and learned linear banks, and TASS yields consistent positive transfer. The framework provides compact spectral modules that plug into graph signal processing pipelines and graph neural networks, combining scalability, interpretability, and cross-graph generalization.

cs.LG↗

arXiv2025

JaxMARL-HFT: GPU-Accelerated Large-Scale Multi-Agent Reinforcement Learning for High-Frequency Trading

Agent-based modelling (ABM) approaches for high-frequency financial markets are difficult to calibrate and validate, partly due to the large parameter space created by defining fixed agent policies. Multi-agent reinforcement learning (MARL) enables more realistic agent behaviour and reduces the number of free parameters, but the heavy computational cost has so far limited research efforts. To address this, we introduce JaxMARL-HFT (JAX-based Multi-Agent Reinforcement Learning for High-Frequency Trading), the first GPU-accelerated open-source multi-agent reinforcement learning environment for high-frequency trading (HFT) on market-by-order (MBO) data. Extending the JaxMARL framework and building on the JAX-LOB implementation, JaxMARL-HFT is designed to handle a heterogeneous set of agents, enabling diverse observation/action spaces and reward functions. It is designed flexibly, so it can also be used for single-agent RL, or extended to act as an ABM with fixed-policy agents. Leveraging JAX enables up to a 240x reduction in end-to-end training time, compared with state-of-the-art reference implementations on the same hardware. This significant speed-up makes it feasible to exploit the large, granular datasets available in high-frequency trading, and to perform the extensive hyperparameter sweeps required for robust and efficient MARL research in trading. We demonstrate the use of JaxMARL-HFT with independent Proximal Policy Optimization (IPPO) for a two-player environment, with an order execution and a market making agent, using one year of LOB data (400 million orders), and show that these agents learn to outperform standard benchmarks. The code for the JaxMARL-HFT framework is available on GitHub.

q-fin.TR↗

arXiv2025

DeltaLag: Learning Dynamic Lead-Lag Patterns in Financial Markets

The lead-lag effect, where the price movement of one asset systematically precedes that of another, has been widely observed in financial markets and conveys valuable predictive signals for trading. However, traditional lead-lag detection methods are limited by their reliance on statistical analysis methods and by the assumption of persistent lead-lag patterns, which are often invalid in dynamic market conditions. In this paper, we propose \textbf{DeltaLag}, the first end-to-end deep learning method that discovers and exploits dynamic lead-lag structures with pair-specific lag values in financial markets for portfolio construction. Specifically, DeltaLag employs a sparsified cross-attention mechanism to identify relevant lead-lag pairs. These lead-lag signals are then leveraged to extract lag-aligned raw features from the leading stocks for predicting the lagger stock's future return. Empirical evaluations show that DeltaLag substantially outperforms both fixed-lag and self-lead-lag baselines. In addition, its adaptive mechanism for identifying lead-lag relationships consistently surpasses precomputed lead-lag graphs based on statistical methods. Furthermore, DeltaLag outperforms a wide range of temporal and spatio-temporal deep learning models designed for stock prediction or time series forecasting, offering both better trading performance and enhanced interpretability.

cs.CE↗

arXiv2025

Factor-Driven Network Informed Restricted Vector Autoregression

High-dimensional financial time series often exhibit complex dependence relations driven by both common market structures and latent connections among assets. To capture these characteristics, this paper proposes Factor-Driven Network Informed Restricted Vector Autoregression (FNIRVAR), a model for the common and idiosyncratic components of high-dimensional time series with an underlying unobserved network structure. The common component is modelled by a static factor model, which allows for strong cross-sectional dependence, whilst a network vector autoregressive process captures the residual co-movements due to the idiosyncratic component. An assortative stochastic block model underlies the network VAR, leading to groups of highly co-moving variables in the idiosyncratic component. For estimation, a two-step procedure is proposed, whereby the static factors are estimated via principal component analysis, followed by estimation of the network VAR parameters. The method is demonstrated in financial applications to daily returns, intraday returns, and FRED-MD macroeconomic variables. In all cases, the proposed method outperforms a static factor model, as well as a static factor plus LASSO-estimated sparse VAR model, in terms of forecasting and financial performance metrics.

stat.ME↗

arXiv2025

On the Stability of Graph Convolutional Neural Networks: A Probabilistic Perspective

Graph convolutional neural networks (GCNNs) have emerged as powerful tools for analyzing graph-structured data, achieving remarkable success across diverse applications. However, the theoretical understanding of the stability of these models, i.e., their sensitivity to small changes in the graph structure, remains in rather limited settings, hampering the development and deployment of robust and trustworthy models in practice. To fill this gap, we study how perturbations in the graph topology affect GCNN outputs and propose a novel formulation for analyzing model stability. Unlike prior studies that focus only on worst-case perturbations, our distribution-aware formulation characterizes output perturbations across a broad range of input data. This way, our framework enables, for the first time, a probabilistic perspective on the interplay between the statistical properties of the node data and perturbations in the graph topology. We conduct extensive experiments to validate our theoretical findings and demonstrate their benefits over existing baselines, in terms of both representation stability and adversarial attacks on downstream tasks. Our results demonstrate the practical significance of the proposed formulation and highlight the importance of incorporating data distribution into stability analysis.

cs.LG↗

arXiv2025

Forecasting Intraday Volume in Equity Markets with Machine Learning

This study focuses on forecasting intraday trading volumes, a crucial component for portfolio implementation, especially in high-frequency (HF) trading environments. Given the current scarcity of flexible methods in this area, we employ a suite of machine learning (ML) models enriched with numerous HF predictors to enhance the predictability of intraday trading volumes. Our findings reveal that intraday stock trading volume is highly predictable, especially with ML and considering commonality. Additionally, we assess the economic benefits of accurate volume forecasting through Volume Weighted Average Price (VWAP) strategies. The results demonstrate that precise intraday forecasting offers substantial advantages, providing valuable insights for traders to optimize their strategies.

q-fin.CP↗

arXiv2025

Can LLM-based Financial Investing Strategies Outperform the Market in Long Run?

Large Language Models (LLMs) have recently been leveraged for asset pricing tasks and stock trading applications, enabling AI agents to generate investment decisions from unstructured financial data. However, most evaluations of LLM timing-based investing strategies are conducted on narrow timeframes and limited stock universes, overstating effectiveness due to survivorship and data-snooping biases. We critically assess their generalizability and robustness by proposing FINSABER, a backtesting framework evaluating timing-based strategies across longer periods and a larger universe of symbols. Systematic backtests over two decades and 100+ symbols reveal that previously reported LLM advantages deteriorate significantly under broader cross-section and over a longer-term evaluation. Our market regime analysis further demonstrates that LLM strategies are overly conservative in bull markets, underperforming passive benchmarks, and overly aggressive in bear markets, incurring heavy losses. These findings highlight the need to develop LLM strategies that are able to prioritise trend detection and regime-aware risk controls over mere scaling of framework complexity.

q-fin.TR↗

arXiv2025

ClusterLOB: Enhancing Trading Strategies by Clustering Orders in Limit Order Books

In the rapidly evolving world of financial markets, understanding the dynamics of limit order book (LOB) is crucial for unraveling market microstructure and participant behavior. We introduce ClusterLOB as a method to cluster individual market events in a stream of market-by-order (MBO) data into different groups. To do so, each market event is augmented with six time-dependent features. By applying the K-means++ clustering algorithm to the resulting order features, we are then able to assign each new order to one of three distinct clusters, which we identify as directional, opportunistic, and market-making participants, each capturing unique trading behaviors. Our experimental results are performed on one year of MBO data containing small-tick, medium-tick, and large-tick stocks from NASDAQ. To validate the usefulness of our clustering, we compute order flow imbalances across each cluster within 30-minute buckets during the trading day. We treat each cluster's imbalance as a signal that provides insights into trading strategies and participants' responses to varying market conditions. To assess the effectiveness of these signals, we identify the trading strategy with the highest Sharpe ratio in the training dataset, and demonstrate that its performance in the test dataset is superior to benchmark trading strategies that do not incorporate clustering. We also evaluate trading strategies based on order flow imbalance decompositions across different market event types, including add, cancel, and trade events, to assess their robustness in various market conditions. This work establishes a robust framework for clustering market participant behavior, which helps us to better understand market microstructure, and inform the development of more effective predictive trading signals with practical applications in algorithmic trading and quantitative finance.

q-fin.TR↗

arXiv2025

Tactical Asset Allocation with Macroeconomic Regime Detection

This paper extends the tactical asset allocation literature by incorporating regime modeling using techniques from machine learning. We propose a novel model that classifies current regimes, forecasts the distribution of future regimes, and integrates these forecasts with the historical performance of individual assets to optimize portfolio allocations. Utilizing a macroeconomic data set from the FRED-MD database, our approach employs a modified k-means algorithm to ensure consistent regime classification over time. We then leverage these regime predictions to estimate expected returns and volatilities, which are subsequently mapped into portfolio allocations using various sizing schemes. Our method outperforms traditional benchmarks such as equal-weight, buy-and-hold, and random regime models. Additionally, we are the first to apply a regime detection model from a large macroeconomic dataset to tactical asset allocation, demonstrating significant improvements in portfolio performance. Our work presents several key contributions, including a novel data-driven regime detection algorithm tailored for uncertainty in forecasted regimes and applying the FRED-MD data set for tactical asset allocation.

q-fin.PM↗

arXiv2025

The Market Maker's Dilemma: Navigating the Fill Probability vs. Post-Fill Returns Trade-Off

Using data from a live trading experiment on the Binance Bitcoin perpetual, we examine the effects of (i) basic order book mechanics and (ii) the persistence of price changes from immediate to short timescales, revealing the interplay between returns, queue sizes, and orders' queue positions. We document a fundamental trade-off: a negative correlation between maker fill likelihood and post-fill returns. This dictates that viable maker strategies often require a contrarian approach, counter-trading the prevailing order book imbalance. These dynamics render commonly-cited strategies highly unprofitable, leading us to model `Reversals': situations where a contrarian maker strategy at the touch proves effective.

q-fin.TR↗

arXiv2025

Generalized Factor Neural Network Model for High-dimensional Regression

We tackle the challenges of modeling high-dimensional data sets, particularly those with latent low-dimensional structures hidden within complex, non-linear, and noisy relationships. Our approach enables a seamless integration of concepts from non-parametric regression, factor models, and neural networks for high-dimensional regression. Our approach introduces PCA and Soft PCA layers, which can be embedded at any stage of a neural network architecture, allowing the model to alternate between factor modeling and non-linear transformations. This flexibility makes our method especially effective for processing hierarchical compositional data. We explore ours and other techniques for imposing low-rank structures on neural networks and examine how architectural design impacts model performance. The effectiveness of our method is demonstrated through simulation studies, as well as applications to forecasting future price movements of equity ETF indices and nowcasting with macroeconomic data.

stat.ML↗

arXiv2025

Learning to Fuse Temporal Proximity Networks: A Case Study in Chimpanzee Social Interactions

How can we identify groups of primate individuals which could be conjectured to drive social structure? To address this question, one of us has collected a time series of data for social interactions between chimpanzees. Here we use a network representation, leading to the task of combining these data into a time series of a single weighted network per time stamp, where different proximities should be given different weights reflecting their relative importance. We optimize these proximity-type weights in a principled way, using an innovative loss function which rewards structural consistency for consecutive time steps. The approach is empirically validated by carefully designed synthetic data. Using statistical tests, we provide a way of identifying groups of individuals that stay related for a significant length of time. Applying the approach to the chimpanzee data set, we detect cliques in the animal social network time series, which can be validated by real-world intuition from prior research and qualitative observations by chimpanzee experts.

stat.ML↗

arXiv2024

GDP nowcasting with large-scale inter-industry payment data in real time -- A network approach

Real-time economic information is essential for policy-making but difficult to obtain. We introduce a granular nowcasting method for macro- and industry-level GDP using a network approach and data on real-time monthly inter-industry payments in the UK. To this purpose we devise a model which we call an extended generalised network autoregressive (GNAR-ex) model, tailored for networks with time-varying edge weights and nodal time series, that exploits the notion of neighbouring nodes and neighbouring edges. The performance of the model is illustrated on a range of synthetic data experiments. We implement the GNAR-ex model on the payments network including time series information of GDP and payment amounts. To obtain robustness against statistical revisions, we optimise the model over 9 quarterly releases of GDP data from the UK Office for National Statistics. Our GNAR-ex model can outperform baseline autoregressive benchmark models, leading to a reduced forecasting error. This work helps to obtain timely GDP estimates at the aggregate and industry level derived from alternative data sources compared to existing, mostly survey-based, methods. Thus, this paper contributes both, a novel model for networks with nodal time series and time-varying edge weights, and the first network-based approach for GDP nowcasting based on payments data.

stat.AP↗