Topic overview

q-fin.ST

279 works747 researchers0 institutions

Topic snapshot

What this area looks like now

279works

747authors

0experts visible

0communities

Next steps

Move from topic reading into action

The graph preview below keeps the nearby papers, people and communities visible in the same reading flow.

Inspect nearby papers, researchers, institutions and communities without opening a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2023arXiv

Pricing European Options under Stochastic Volatility Models: Case of five-Parameter Variance-Gamma Process

The paper builds a Variance-Gamma (VG) model with five parameters: location ($μ$), symmetry ($δ$), volatility ($σ$), shape ($α$), and scale ($θ$); and studies its application to the pricing of European options. The results of our analysis show that the five-parameter VG model is a stochastic volatility model with a $Γ(α, θ)$ Ornstein-Uhlenbeck type process; the associated Lévy density of the VG model is a KoBoL family of order $ν=0$, intensity $α$, and steepness parameters $\fracδ{σ^2} - \sqrt{\frac{δ^2}{σ^4}+\frac{2}{θσ^2}}$ and $\fracδ{σ^2}+ \sqrt{\frac{δ^2}{σ^4}+\frac{2}{θσ^2}}$; and the VG process converges asymptotically in distribution to a Lévy process driven by a normal distribution with mean $(μ+ αθδ)$ and variance $α(θ^2δ^2 + σ^2θ)$. The data used for empirical analysis were obtained by fitting the five-parameter Variance-Gamma (VG) model to the underlying distribution of the daily SPY ETF data. Regarding the application of the five-parameter VG model, the twelve-point rule Composite Newton-Cotes Quadrature and Fractional Fast Fourier (FRFT) algorithms were implemented to compute the European option price. Compared to the Black-Scholes (BS) model, empirical evidence shows that the VG option price is underpriced for out-of-the-money (OTM) options and overpriced for in-the-money (ITM) options. Both models produce almost the same option pricing results for deep out-of-the-money (OTM) and deep-in-the-money (ITM) options

preprint2026arXiv

Vector-Quantized Discrete Latent Factors Meet Financial Priors: Dynamic Cross-Sectional Stock Ranking Prediction for Portfolio Construction

Predicting cross-sectional stock returns is challenging due to low signal-to-noise ratios and evolving market regimes. Classical factor models offer interpretability but limited flexibility, while deep learning models achieve strong performance yet often underutilize financial priors. We address this gap with PRISM-VQ (PRior-Informed Stock Model with Vector Quantization), a dynamic factor framework that integrates expert prior factors, vector-quantized discrete latent factors learned from cross-sectional structure, and a structure-conditioned Mixture-of-Experts to generate time-varying factor loadings. Vector quantization acts as an information bottleneck that suppresses noise while capturing robust market structure, with discrete codes serving both as latent factors and as routing signals for temporal expert specialization. Experiments on CSI 300 and S&P 500 show consistent improvements in cross-sectional return prediction and portfolio performance over strong baselines while preserving interpretability. Our code is available at https://github.com/finxlab/PRISM-VQ.

preprint2026arXiv

Enhancing a Risk Model by Adding Transient Statistical Factors

Estimating the covariance of asset returns, i.e., the risk model, is a key component of financial portfolio construction and evaluation. Most risk modeling approaches produce a factor model that decomposes the asset variability into two components: the first attributed to a small number of factors that are common among the assets and the second attributed to the idiosyncratic behavior of each asset. Third-party providers typically provide risk models to investors, and while these models are typically of high quality, they may fail to capture important information, e.g., changing market regimes and transient factors. To overcome these limitations, we propose a systematic method based on maximum likelihood estimation to enhance an existing factor model by both refining the given model and adding new statistical factors. Our approach relies only on the observed sequence of realized returns and on the choice of two hyperparameters: the number of additional factors and the half-life parameter that determines the weights assigned to returns in the log-likelihood objective. Importantly, our methodology applies to the situation where asset returns may be missing, making it suitable for typical equity datasets. We demonstrate our approach on the Barra short-term US risk model, a high-quality risk model used in practice, for a universe of US high-capitalization equities. We show that the proposed extension captures structure in the returns that is missed by the original model.

preprint2026arXiv

Bi-Level Chaotic Fusion Based Graph Convolutional Network for Stock Market Prediction Interval

Financial market forecasting is inherently uncertain, yet most deep learning approaches rely on point predictions that provide only single-value estimates without quantifying uncertainty. Such predictions are insufficient for risk-aware decision-making, as they fail to capture the range of possible outcomes and the associated confidence of forecasts.The problem can be solved using prediction intervals, which allow obtaining an upper and lower bound for the prediction, thus enabling uncertainty representation in the model. Yet, the current methods tend to disregard relationships between assets or cannot simultaneously ensure good calibration and sharpness of the resulting intervals in dynamically changing market regimes. In our work, we propose a spatio-temporal graph-based approach with a bi-level chaotic fusion technique to solve this problem. Our model uses separate nonlinear transformation functions to estimate the interval center and width. Additionally, a volatility-aware gating mechanism is used to make predictions dependent on the regime in which the market operates. Temporal dependencies are considered by embedding graph structures and sequentially modeling them. Training is conducted according to a Lower-Upper Bound Estimation (LUBE) objective. Our experimental results show significant improvements compared to existing baselines (LSTM, GRU, GCN, HGNN) when applied to data from 2016 to 2026 with 43 leading companies in eight sectors of the NSE. It provides the lowest Winkler score (0.0778), tightest prediction intervals (PIAW = 0.1407), and highest coverage (PICP = 96.6%), with all differences statistically significant (p < 0.001) according to the Diebold-Mariano test.

preprint2022arXiv

Cross-Domain Shopping and Stock Trend Analysis

This paper presents a cross-domain trend analysis that aims to identify and analyze the relationships between stock prices, stock news on Twitter, and users' behaviors on e-commerce websites. The analysis is based on three datasets: a US stock dataset, a stock tweets dataset, and an e-commerce behavior dataset. The analysis is performed using Hadoop, Hive, and Tableau, allowing for efficient and scalable processing and visualizing large datasets. The analysis includes trend analysis of Twitter sentiment (positive and negative tweets) and correlation analysis, including the correlation between tweet sentiment and stocks, the correlation between stock trends and shopping behavior, and the understanding of data based on different slices of time. By comparing different features from the datasets over time, we hope to gain insight into the factors that drive user behavior as well as the market in different categories. The results of this analysis can provide valuable insights for businesses and investors to inform decision-making. We believe that our analysis can serve as a valuable starting point for further research and investigation into these topics.

preprint2026arXiv

Statistical Model Checking of the Keynes+Schumpeter Model: A Transient Sensitivity Analysis of a Macroeconomic ABM

Agent-based models (ABMs) are increasingly used in macroeconomics, but their analysis still often relies on ad hoc Monte Carlo campaigns with heterogeneous statistical effort across parameter settings. We show how statistical model checking (SMC), implemented through MultiVeStA, can provide a principled analysis layer for a realistic macroeconomic ABM without rewriting the simulator in a dedicated formalism. Our case study is the heuristic-switching Keynes+Schumpeter(K+S) model, analysed hrough a transient sensitivity campaign over one-parameter sweeps, two macro observables (unemployment and GDP growth), and one auxiliary micro-level probe (market share) on the post-warmup phase of a 600-step horizon. The analysis is driven by reusable temporal queries, observable-specific precision targets, and confidence-based stopping rules that automatically determine the simulation effort required by each configuration. Results show a clear contrast across parameter families: macro-financial and structural sweeps produce the strongest transient effects, whereas several heuristic-rule sweeps remain much weaker under the same precision policy. More broadly, the paper shows that SMC can support reproducible and informative quantitative analysis of substantively rich economic ABMs, while making uncertainty estimates and simulation cost explicit parts of the reported results.

preprint2022arXiv

Homogeneity and heterogeneity of cryptocurrencies

Thousands of cryptocurrencies have been issued and publicly exchanged since Bitcoin was invented in 2008. The total cryptocurrency market value exceeds 300 billion US dollars as of 2019. This paper analyzes the prices, volumes, blockchain transactions, coin difficulties and public opinion popularities of 3607 actively exchanged cryptocurrencies. We aim to reveal and explain the homogeneity, i.e., the strong correlation of market performance, and the heterogeneity, i.e., the imbalance of popularities and sophistications, of the cryptocurrencies.

preprint2014arXiv

Efficient Modeling and Forecasting of the Electricity Spot Price

The increasing importance of renewable energy, especially solar and wind power, has led to new forces in the formation of electricity prices. Hence, this paper introduces an econometric model for the hourly time series of electricity prices of the European Power Exchange (EPEX) which incorporates specific features like renewable energy. The model consists of several sophisticated and established approaches and can be regarded as a periodic VAR-TARCH with wind power, solar power, and load as influences on the time series. It is able to map the distinct and well-known features of electricity prices in Germany. An efficient iteratively reweighted lasso approach is used for the estimation. Moreover, it is shown that several existing models are outperformed by the procedure developed in this paper.

preprint2025arXiv

Investigating Conditional Restricted Boltzmann Machines in Regime Detection

This study investigates the efficacy of Conditional Restricted Boltzmann Machines (CRBMs) for modeling high-dimensional financial time series and detecting systemic risk regimes. We extend the classical application of static Restricted Boltzmann Machines (RBMs) by incorporating autoregressive conditioning and utilizing Persistent Contrastive Divergence (PCD) to incorporate complex temporal dependency structures. Comparing a discrete Bernoulli-Bernoulli architecture against a continuous Gaussian-Bernoulli variant across a multi-asset dataset spanning 2013-2025, we observe a dichotomy between generative fidelity and regime detection. While the Gaussian CRBM successfully preserves static asset correlations, it exhibits limitations in generating long-range volatility clustering. Thus, we analyze the free energy as a relative negative log-likelihood (surprisal) under a fixed, trained model. We demonstrate that the model's free energy serves as a robust, regime stability metric. By decomposing the free energy into quadratic (magnitude) and structural (correlation) components, we show that the model can distinguish between pure magnitude shocks and market regimes. Our findings suggest that the CRBM offers a valuable, interpretable diagnostic tool for monitoring systemic risk, providing a supplemental metric to implied volatility metrics like the VIX.

preprint2026arXiv

The Payment Heterogeneity Index: An Integrated Unsupervised Framework for High-Volume Procurement Oversight and Decision Support

Public procurement is vulnerable to error, fraud, and corruption, particularly as high transaction volumes overwhelm oversight. While research often focuses on tender-stage anomalies, post-award payment monitoring remains underexplored. Since labelled datasets are rare and methods like Benford's Law face restrictive assumptions, there is a need for interpretable, unsupervised frameworks for high-volume procurement oversight and decision support. This paper introduces the Structural Heterogeneity Index (SHI), a composite statistic for one-dimensional samples, and its payment-specific instantiation, the Payment Heterogeneity Index (PHI), characterising payment structure and latent regimes. It incorporates Gaussian Mixture Model (GMM) parameters alongside non-parametric statistics, integrating four interpretable components: modality, asymmetry, tail behaviour, and structural dispersion. Uniquely, the tail-behaviour component captures both distributional heaviness and extreme-value concentration, while structural-dispersion combines the variability, prevalence, and separation of latent payment regimes. Applied to UK municipal procurement data, PHI identifies a financially significant cohort (0.6\% of suppliers; 10.1\% of high-volume vendors) with structurally distinct payment patterns. Statistical testing further supports these differences, and targeted human verification confirms the plausibility of prioritised cases. Comparative analysis shows PHI reveals regime separation obscured by the Coefficient of Variation ($ρ= 0.310$). PHI provides a transparent, decomposable, and computationally lightweight framework for procurement integrity oversight and targeted audit prioritisation.

preprint2022arXiv

Kernel Estimation of Spot Volatility with Microstructure Noise Using Pre-Averaging

We first revisit the problem of estimating the spot volatility of an Itô semimartingale using a kernel estimator. We prove a Central Limit Theorem with optimal convergence rate for a general two-sided kernel. Next, we introduce a new pre-averaging/kernel estimator for spot volatility to handle the microstructure noise of ultra high-frequency observations. We prove a Central Limit Theorem for the estimation error with an optimal rate and study the optimal selection of the bandwidth and kernel functions. We show that the pre-averaging/kernel estimator's asymptotic variance is minimal for exponential kernels, hence, justifying the need of working with kernels of unbounded support as proposed in this work. We also develop a feasible implementation of the proposed estimators with optimal bandwidth. Monte Carlo experiments confirm the superior performance of the devised method.

preprint2022arXiv

Time is limited on the road to asymptopia

One challenge in the estimation of financial market agent-based models (FABMs) is to infer reliable insights using numerical simulations validated by only a single observed time series. Ergodicity (besides stationarity) is a strong precondition for any estimation, however it has not been systematically explored and is often simply presumed. For finite-sample lengths and limited computational resources empirical estimation always takes place in pre-asymptopia. Thus broken ergodicity must be considered the rule, but it remains largely unclear how to deal with the remaining uncertainty in non-ergodic observables. Here we show how an understanding of the ergodic properties of moment functions can help to improve the estimation of (F)ABMs. We run Monte Carlo experiments and study the convergence behaviour of moment functions of two prototype models. We find infeasibly-long convergence times for most. Choosing an efficient mix of ensemble size and simulated time length guided our estimation and might help in general.

preprint2022arXiv

New drugs and stock market: how to predict pharma market reaction to clinical trial announcements

Pharmaceutical companies operate in a strictly regulated and highly risky environment in which a single slip can lead to serious financial implications. Accordingly, the announcements of clinical trial results tend to determine the future course of events, hence being closely monitored by the public. In this work, we provide statistical evidence for the result promulgation influence on the public pharma market value. Whereas most works focus on retrospective impact analysis, the present research aims to predict the numerical values of announcement-induced changes in stock prices. For this purpose, we develop a pipeline that includes a BERT-based model for extracting sentiment polarity of announcements, a Temporal Fusion Transformer for forecasting the expected return, a graph convolution network for capturing event relationships, and gradient boosting for predicting the price change. The challenge of the problem lies in inherently different patterns of responses to positive and negative announcements, reflected in a stronger and more pronounced reaction to the negative news. Moreover, such phenomenon as the drop in stocks after the positive announcements affirms the counterintuitiveness of the price behavior. Importantly, we discover two crucial factors that should be considered while working within a predictive framework. The first factor is the drug portfolio size of the company, indicating the greater susceptibility to an announcement in the case of small drug diversification. The second one is the network effect of the events related to the same company or nosology. All findings and insights are gained on the basis of one of the biggest FDA (the Food and Drug Administration) announcement datasets, consisting of 5436 clinical trial announcements from 681 companies over the last five years.

preprint2026arXiv

Sequential Structure in Intraday Futures Data: LSTM vs Gradient Boosting on MNQ

This paper compares gradient boosting and long short-term memory (LSTM) architectures for intraday directional prediction in Micro E-Mini Nasdaq 100 futures (MNQ). Motivated by recent foundation-model research on financial candlestick data, including the Kronos architecture, we test whether five-minute OHLCV bar sequences contain exploitable sequential predictive structure at the scale of a single instrument dataset. Using 944 trading days from 2021-2025, four model configurations are evaluated under strict expanding-window walk-forward validation across three out-of-sample periods. The target variable is whether the session close exceeds the 10:30 AM open by more than ten points. No configuration produces statistically significant out-of-sample accuracy above the 51.8% base rate. Combined OOS accuracies range from 50.00% to 50.89% across gradient boosting variants, while the LSTM achieves 50.59%. Permutation tests yield p-values of 0.135 for the best gradient boosting model and 0.515 for the LSTM, indicating no statistically significant predictive edge. Feature importance instability across walk-forward folds suggests noise fitting rather than stable structural signal capture. The results indicate that four years of single-instrument five-minute OHLCV data are insufficient for reliable sequential ML-based intraday forecasting. The primary contribution is a documented evaluation of a Kronos-inspired architecture on a constrained real-world dataset, providing an empirical lower bound on data scale requirements for sequential financial ML.

preprint2022arXiv

Understanding Unfairness in Fraud Detection through Model and Data Bias Interactions

In recent years, machine learning algorithms have become ubiquitous in a multitude of high-stakes decision-making applications. The unparalleled ability of machine learning algorithms to learn patterns from data also enables them to incorporate biases embedded within. A biased model can then make decisions that disproportionately harm certain groups in society -- limiting their access to financial services, for example. The awareness of this problem has given rise to the field of Fair ML, which focuses on studying, measuring, and mitigating unfairness in algorithmic prediction, with respect to a set of protected groups (e.g., race or gender). However, the underlying causes for algorithmic unfairness still remain elusive, with researchers divided between blaming either the ML algorithms or the data they are trained on. In this work, we maintain that algorithmic unfairness stems from interactions between models and biases in the data, rather than from isolated contributions of either of them. To this end, we propose a taxonomy to characterize data bias and we study a set of hypotheses regarding the fairness-accuracy trade-offs that fairness-blind ML algorithms exhibit under different data bias settings. On our real-world account-opening fraud use case, we find that each setting entails specific trade-offs, affecting fairness in expected value and variance -- the latter often going unnoticed. Moreover, we show how algorithms compare differently in terms of accuracy and fairness, depending on the biases affecting the data. Finally, we note that under specific data bias conditions, simple pre-processing interventions can successfully balance group-wise error rates, while the same techniques fail in more complex settings.

preprint2024arXiv

Structured factor copulas for modeling the systemic risk of European and United States banks

In this paper, we employ Credit Default Swaps (CDS) to model the joint and conditional distress probabilities of banks in Europe and the U.S. using factor copulas. We propose multi-factor, structured factor, and factor-vine models where the banks in the sample are clustered according to their geographic location. We find that within each region, the co-dependence between banks is best described using both, systematic and idiosyncratic, financial contagion channels. However, if we consider the banking system as a whole, then the systematic contagion channel prevails, meaning that the distress probabilities are driven by a latent global factor and region-specific factors. In all cases, the co-dependence structure of bank CDS spreads is highly correlated in the tail. The out-of-sample forecasts of several measures of systematic risk allow us to identify the periods of distress in the banking sector over the recent years including the COVID-19 pandemic, the interest rate hikes in 2022, and the banking crisis in 2023.

preprint2024arXiv

Economic Forces in Stock Returns

When analyzing the components influencing the stock prices, it is commonly believed that economic activities play an important role. More specifically, asset prices are more sensitive to the systematic economic news that impose a pervasive effect on the whole market. Moreover, the investors will not be rewarded for bearing idiosyncratic risks as such risks are diversifiable. In the paper Economic Forces and the Stock Market 1986, the authors introduced an attribution model to identify the specific systematic economic forces influencing the market. They first defined and examined five classic factors from previous research papers: Industrial Production, Unanticipated Inflation, Change in Expected Inflation, Risk Premia, and The Term Structure. By adding in new factors, the Market Indices, Consumptions and Oil Prices, one by one, they examined the significant contribution of each factor to the stock return. The paper concluded that the stock returns are exposed to the systematic economic news, and they are priced with respect to their risk exposure. Also, the significant factors can be identified by simply adopting their model. Driven by such motivation, we conduct an attribution analysis based on the general framework of their model to further prove the importance of the economic factors and identify the specific identity of significant factors.

preprint2022arXiv

Hierarchical Deep Reinforcement Learning for VWAP Strategy Optimization

Designing an intelligent volume-weighted average price (VWAP) strategy is a critical concern for brokers, since traditional rule-based strategies are relatively static that cannot achieve a lower transaction cost in a dynamic market. Many studies have tried to minimize the cost via reinforcement learning, but there are bottlenecks in improvement, especially for long-duration strategies such as the VWAP strategy. To address this issue, we propose a deep learning and hierarchical reinforcement learning jointed architecture termed Macro-Meta-Micro Trader (M3T) to capture market patterns and execute orders from different temporal scales. The Macro Trader first allocates a parent order into tranches based on volume profiles as the traditional VWAP strategy does, but a long short-term memory neural network is used to improve the forecasting accuracy. Then the Meta Trader selects a short-term subgoal appropriate to instant liquidity within each tranche to form a mini-tranche. The Micro Trader consequently extracts the instant market state and fulfils the subgoal with the lowest transaction cost. Our experiments over stocks listed on the Shanghai stock exchange demonstrate that our approach outperforms baselines in terms of VWAP slippage, with an average cost saving of 1.16 base points compared to the optimal baseline.

preprint2022arXiv

Variations on two-parameter families of forecasting functions: seasonal/nonseasonal Models, comparison to the exponential smoothing and ARIMA models, and applications to stock market data

We introduce twenty four two-parameter families of advanced time series forecasting functions using a new and nonparametric approach. We also introduce the concept of powering and derive nonseasonal and seasonal models with examples in education, sales, finance and economy. We compare the performance of our twenty four models to both Holt--Winters and ARIMA models for both nonseasonal and seasonal times series. We show in particular that our models not only do not require a decomposition of a seasonal time series into trend, seasonal and random components, but leads also to substantially lower sum of absolute error and a higher number of closer forecasts than both Holt--Winters and ARIMA models. Finally, we apply and compare the performance of our twenty four models using five-year stock market data of 467 companies of the S&P500.

preprint2023arXiv

Financial Time-Series Forecasting: Towards Synergizing Performance And Interpretability Within a Hybrid Machine Learning Approach

In the realm of cryptocurrency, the prediction of Bitcoin prices has garnered substantial attention due to its potential impact on financial markets and investment strategies. This paper propose a comparative study on hybrid machine learning algorithms and leverage on enhancing model interpretability. Specifically, linear regression(OLS, LASSO), long-short term memory(LSTM), decision tree regressors are introduced. Through the grounded experiments, we observe linear regressor achieves the best performance among candidate models. For the interpretability, we carry out a systematic overview on the preprocessing techniques of time-series statistics, including decomposition, auto-correlational function, exponential triple forecasting, which aim to excavate latent relations and complex patterns appeared in the financial time-series forecasting. We believe this work may derive more attention and inspire more researches in the realm of time-series analysis and its realistic applications.

preprint2023arXiv

Cluster-based Regression using Variational Inference and Applications in Financial Forecasting

This paper describes an approach to simultaneously identify clusters and estimate cluster-specific regression parameters from the given data. Such an approach can be useful in learning the relationship between input and output when the regression parameters for estimating output are different in different regions of the input space. Variational Inference (VI), a machine learning approach to obtain posterior probability densities using optimization techniques, is used to identify clusters of explanatory variables and regression parameters for each cluster. From these results, one can obtain both the expected value and the full distribution of predicted output. Other advantages of the proposed approach include the elegant theoretical solution and clear interpretability of results. The proposed approach is well-suited for financial forecasting where markets have different regimes (or clusters) with different patterns and correlations of market changes in each regime. In financial applications, knowledge about such clusters can provide useful insights about portfolio performance and identify the relative importance of variables in different market regimes. An illustrative example of predicting one-day S&P change is considered to illustrate the approach and compare the performance of the proposed approach with standard regression without clusters. Due to the broad applicability of the problem, its elegant theoretical solution, and the computational efficiency of the proposed algorithm, the approach may be useful in a number of areas extending beyond the financial domain.

preprint2022arXiv

Fitting Generalized Tempered Stable distribution: Fractional Fourier Transform (FRFT) Approach

The paper investigates the rich class of Generalized Tempered Stable distribution, an alternative to Normal distribution and the $α$-Stable distribution for modelling asset return and many physical and economic systems. Firstly, we explore some important properties of the Generalized Tempered Stable (GTS) distribution. The theoretical tools developed are used to perform empirical analysis. The GTS distribution is fitted using S&P 500, SPY ETF and Bitcoin BTC. The Fractional Fourier Transform (FRFT) technique evaluates the probability density function and its derivatives in the maximum likelihood procedure. Based on the results from the statistical inference and the Kolmogorov-Smirnov (K-S) goodness-of-fit, the GTS distribution fits the underlying distribution of the SPY ETF return. The right side of the Bitcoin BTC return, and the left side of the S&P 500 return underlying distributions fit the Tempered Stable distribution; while the left side of the Bitcoin BTC return and the right side of the S&P 500 return underlying distributions are modelled by the compound Poisson process

preprint2022arXiv

Stock Prices as Janardan Galton Watson Process

Janardan (1980) introduces a class of offspring distributions that sandwich between Bernoulli and Poisson. This paper extends the Janardan Galton Watson (JGW) branching process as a model of stock prices. In this article, the return value over time t depends on the initial close price, which shows the number of offspring, has a role in the expectation of return and probability of extinction after the passage at time t. Suppose the number of offspring in t th generation is zero, (i.e., called extinction of model at time t) is equivalent with negative return values over time [0, t]. We also introduce the Algorithm that detecting the trend of stock markets.

preprint2026arXiv

GeomHerd: A Forward-looking Herding Quantification via Ricci Flow Geometry on Agent Interactive Simulations

Herding -- where agents align their behaviors and act collectively -- is a central driver of market fragility and systemic risk. Existing approaches to quantify herding rely on price-correlation statistics, which inherently lag because they only detect coordination after it has already moved realised returns. We propose GeomHerd, a forward-looking geometric framework that bypasses this observability lag by quantifying coordination directly on upstream agent-interaction graphs. To generate these graphs, we treat a heterogeneous LLM-driven multi-agent simulator -- each financial trader instantiated by a persona-conditioned LLM call -- as a forecastable world, and evaluate the geometric pipeline on the Cividino--Sornette continuous-spin agent-based substrate as our headline financial testbed. By tracking the discrete Ollivier--Ricci curvature of these action graphs, GeomHerd captures the structural topology of emerging coordination. Theoretically, we establish a mean-field bridge mapping our graph-theoretic metric to CSAD, the classical macroscopic herding statistic, linking GeomHerd to downstream price-dispersion measurement. Empirically, GeomHerd anticipates herding long before aggregate market baselines: on the continuous-spin substrate, our primary detector fires a median of 272 steps before order-parameter onset; a contagion detector ($β_{-}$) recalls 65% of critical trajectories 318 steps early; and on co-firing trajectories the agent-graph signal precedes price-correlation-graph baselines by 40 steps. As a complementary indicator, the effective vocabulary of agent actions contracts during cascades. The geometric signature transfers out-of-domain to the Vicsek self-driven-particle model, and a curvature-conditioned forecasting head reduces cascade-window log-return MAE over detector-conditioned and price-only baselines.