Microstructure and Manipulation: Quantifying Pump-and-Dump Dynamics in Cryptocurrency Markets ArXiv ID: 2504.15790 “View on arXiv” Authors: Unknown Abstract Building on our prior threshold-based analysis of six months of Poloniex trading data, we have extended both the temporal span and granularity of our study by incorporating minute-level OHLCV records for 1021 tokens around each confirmed pump-and-dump event. First, we algorithmically identify the accumulation phase, marking the initial and final insider volume spikes, and observe that 70% of pre-event volume transacts within one hour of the pump announcement. Second, we compute conservative lower bounds on insider profits under both a single-point liquidation at 70% of peak and a tranche-based strategy (selling 20% at 50%, 30% at 60%, and 50% at 80% of peak), yielding median returns above 100% and upper-quartile returns exceeding 2000%. Third, by unfolding the full pump structure and integrating social-media verification (e.g., Telegram announcements), we confirm numerous additional events that eluded our initial model. We also categorize schemes into “pre-accumulation” versus “on-the-spot” archetypes-insights that sharpen detection algorithms, inform risk assessments, and underpin actionable strategies for real-time market-integrity enforcement. ...

Modeling and Forecasting Realized Volatility with Multivariate Fractional Brownian Motion ArXiv ID: 2504.15985 “View on arXiv” Authors: Unknown Abstract A multivariate fractional Brownian motion (mfBm) with component-wise Hurst exponents is used to model and forecast realized volatility. We investigate the interplay between correlation coefficients and Hurst exponents and propose a novel estimation method for all model parameters, establishing consistency and asymptotic normality of the estimators. Additionally, we develop a time-reversibility test, which is typically not rejected by real volatility data. When the data-generating process is a time-reversible mfBm, we derive optimal forecasting formulae and analyze their properties. A key insight is that an mfBm with different Hurst exponents and non-zero correlations can reduce forecasting errors compared to a one-dimensional model. Consistent with optimal forecasting theory, out-of-sample forecasts using the time-reversible mfBm show improvements over univariate fBm, particularly when the estimated Hurst exponents differ significantly. Empirical results demonstrate that mfBm-based forecasts outperform the (vector) HAR model. ...

Realized Local Volatility Surface ArXiv ID: 2504.15626 “View on arXiv” Authors: Unknown Abstract For quantitative trading risk management purposes, we present a novel idea: the realized local volatility surface. Concisely, it stands for the conditional expected volatility when sudden market behaviors of the underlying occur. One is able to explore risk management usages by following the orthotical Delta-Gamma dynamic hedging framework. The realized local volatility surface is, mathematically, a generalized Wiener measure from historical prices. It is reconstructed via employing high-frequency trading market data. A Stick-Breaking Gaussian Mixture Model is fitted via Hamiltonian Monte Carlo, producing a local volatility surface with 95% credible intervals. A practically validated Bayesian nonparametric estimation workflow. Empirical results on TSLA high-frequency data illustrate its ability to capture counterfactual volatility. We also discuss its application in improving volatility-based risk management. ...

Beyond Correlation: Positive Definite Dependence Measures for Robust Inference, Flexible Scenarios, and Causal Modeling for Financial Portfolios ArXiv ID: 2504.15268 “View on arXiv” Authors: Unknown Abstract We live in a multivariate world, and effective modeling of financial portfolios, including their construction, allocation, forecasting, and risk analysis, simply is not possible without explicitly modeling the dependence structure of their assets. Dependence structure can drive portfolio results more than the combined effects of other parameters in investment and risk models, but the literature provides relatively little to define the finite-sample distributions of dependence measures under challenging, real-world financial data conditions. Yet this is exactly what is needed to make valid inferences about their estimates, and to use these inferences for essential purposes such as hypothesis testing, dynamic monitoring, realistic and granular scenario and reverse scenario analyses, and mitigating the effects of correlation breakdowns during market upheavals. This work develops a new and straightforward method, Nonparametric Angles-based Correlation (NAbC), for defining the finite-sample distributions of any dependence measure whose matrix of pairwise associations is positive definite (e.g. Pearsons, Kendalls, Spearmans, the Tail Dependence Matrix, and others). The solution remains valid under marginal asset distributions characterized by notably different and varying degrees of serial correlation, non-stationarity, heavy-tailedness, and asymmetry. Importantly, it provides p-values and confidence intervals at the matrix level, even when selected cells in the matrix are frozen, thus enabling flexible, granular, and realistic scenarios, reverse scenarios, and stress tests. Finally, when applied to directional dependence measures, NAbC enables accurate DAG recovery in causal modeling. NAbC stands alone in providing all of these capabilities simultaneously, and should prove to be a very useful means by which we can better understand and manage financial portfolios in our multivariate world. ...

Deep Reinforcement Learning for Investor-Specific Portfolio Optimization: A Volatility-Guided Asset Selection Approach ArXiv ID: 2505.03760 “View on arXiv” Authors: Unknown Abstract Portfolio optimization requires dynamic allocation of funds by balancing the risk and return tradeoff under dynamic market conditions. With the recent advancements in AI, Deep Reinforcement Learning (DRL) has gained prominence in providing adaptive and scalable strategies for portfolio optimization. However, the success of these strategies depends not only on their ability to adapt to market dynamics but also on the careful pre-selection of assets that influence overall portfolio performance. Incorporating the investor’s preference in pre-selecting assets for a portfolio is essential in refining their investment strategies. This study proposes a volatility-guided DRL-based portfolio optimization framework that dynamically constructs portfolios based on investors’ risk profiles. The Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model is utilized for volatility forecasting of stocks and categorizes them based on their volatility as aggressive, moderate, and conservative. The DRL agent is then employed to learn an optimal investment policy by interacting with the historical market data. The efficacy of the proposed methodology is established using stocks from the Dow $30$ index. The proposed investor-specific DRL-based portfolios outperformed the baseline strategies by generating consistent risk-adjusted returns. ...

The Memorization Problem: Can We Trust LLMs’ Economic Forecasts? ArXiv ID: 2504.14765 “View on arXiv” Authors: Unknown Abstract Large language models (LLMs) cannot be trusted for economic forecasts during periods covered by their training data. Counterfactual forecasting ability is non-identified when the model has seen the realized values: any observed output is consistent with both genuine skill and memorization. Any evidence of memorization represents only a lower bound on encoded knowledge. We demonstrate LLMs have memorized economic and financial data, recalling exact values before their knowledge cutoff. Instructions to respect historical boundaries fail to prevent recall-level accuracy, and masking fails as LLMs reconstruct entities and dates from minimal context. Post-cutoff, we observe no recall. Memorization extends to embeddings. ...

LLM-Enhanced Black-Litterman Portfolio Optimization ArXiv ID: 2504.14345 “View on arXiv” Authors: Unknown Abstract The Black-Litterman model addresses the sensitivity issues of tra- ditional mean-variance optimization by incorporating investor views, but systematically generating these views remains a key challenge. This study proposes and validates a systematic frame- work that translates return forecasts and predictive uncertainty from Large Language Models (LLMs) into the core inputs for the Black-Litterman model: investor views and their confidence lev- els. Through a backtest on S&P 500 constituents, we demonstrate that portfolios driven by top-performing LLMs significantly out- perform traditional baselines in both absolute and risk-adjusted terms. Crucially, our analysis reveals that each LLM exhibits a dis- tinct and consistent investment style which is the primary driver of performance. We found that the selection of an LLM is therefore not a search for a single best forecaster, but a strategic choice of an investment style whose success is contingent on its alignment with the prevailing market regime. The source code and data are available at https://github.com/youngandbin/LLM-BLM. ...

Numerical analysis of a particle system for the calibrated Heston-type local stochastic volatility model ArXiv ID: 2504.14343 “View on arXiv” Authors: Unknown Abstract We analyse a Monte Carlo particle method for the simulation of the calibrated Heston-type local stochastic volatility (H-LSV) model. The common application of a kernel estimator for a conditional expectation in the calibration condition results in a McKean-Vlasov (MV) stochastic differential equation (SDE) with non-standard coefficients. The primary challenges lie in certain mean-field terms in the drift and diffusion coefficients and the $1/2$-Hölder regularity of the diffusion coefficient. We establish the well-posedness of this equation for a fixed but arbitrarily small bandwidth of the kernel estimator. Moreover, we prove a strong propagation of chaos result, ensuring convergence of the particle system under a condition on the Feller ratio and up to a critical time. For the numerical simulation, we employ an Euler-Maruyama scheme for the log-spot process and a full truncation Euler scheme for the CIR volatility process. Under certain conditions on the inputs and the Feller ratio, we prove strong convergence of the Euler-Maruyama scheme with rate $1/2$ in time, up to a logarithmic factor. Numerical experiments illustrate the convergence of the discretisation scheme and validate the propagation of chaos in practice. ...

Cross-Modal Temporal Fusion for Financial Market Forecasting ArXiv ID: 2504.13522 “View on arXiv” Authors: Unknown Abstract Accurate forecasting in financial markets requires integrating diverse data sources, from historical prices to macroeconomic indicators and financial news. However, existing models often fail to align these modalities effectively, limiting their practical use. In this paper, we introduce a transformer-based deep learning framework, Cross-Modal Temporal Fusion (CMTF), that fuses structured and unstructured financial data for improved market prediction. The model incorporates a tensor interpretation module for feature selection and an auto-training pipeline for efficient hyperparameter tuning. Experimental results using FTSE 100 stock data demonstrate that CMTF achieves superior performance in price direction classification compared to classical and deep learning baselines. These findings suggest that our framework is an effective and scalable solution for real-world cross-modal financial forecasting tasks. ...

Deep Learning Models Meet Financial Data Modalities ArXiv ID: 2504.13521 “View on arXiv” Authors: Unknown Abstract Algorithmic trading relies on extracting meaningful signals from diverse financial data sources, including candlestick charts, order statistics on put and canceled orders, traded volume data, limit order books, and news flow. While deep learning has demonstrated remarkable success in processing unstructured data and has significantly advanced natural language processing, its application to structured financial data remains an ongoing challenge. This study investigates the integration of deep learning models with financial data modalities, aiming to enhance predictive performance in trading strategies and portfolio optimization. We present a novel approach to incorporating limit order book analysis into algorithmic trading by developing embedding techniques and treating sequential limit order book snapshots as distinct input channels in an image-based representation. Our methodology for processing limit order book data achieves state-of-the-art performance in high-frequency trading algorithms, underscoring the effectiveness of deep learning in financial applications. ...