Volatility dynamics can carry information about temporal relationships that is not apparent in the conditional mean. Extracting this information is difficult because the persistence of each variable c...
arXiv:2609.06085v1 Announce Type: cross Abstract: Understanding the joint dynamics of prices and trades is central to market microstructure, where returns and order flow interact through nonlinear an...
arXiv:2610.07430v1 Announce Type: new Abstract: Reducing nonstationarity in a persistent time series entails deciding how much of its temporal dependence to remove. In finance, fractional differencin...
The paper presents a Causal Graph‑Informed Temporal Convolutional Network (CG‑TCN) that fuses a learned causal graph with a temporal convolutional network to forecast retail electricity prices. By decomposing price series into multi‑resolution trends and discovering a causal graph over these components and key covariates, the model conditions its convolutions and attention on causal pathways. On ten years of Ohio residential contracts, CG‑TCN outperforms benchmarks, achieving MAEps of 3.08%, 3.82%, and 5.43% for one‑, ten‑, and fifteen‑step‑ahead forecasts, respectively.
VertiFuseX is a hybrid LSTM architecture that fuses multi‑scale temporal representations at the penultimate layer, stacking features from LSTM, Bi‑LSTM, and St‑LSTM branches and a parallel DNN stream. On 15 years of global equity index data, it reduces MAPE by 30‑54% and improves MAE and RMSE by over 40% compared to LSTM baselines, outperforming seven state‑of‑the‑art models across 33 metric‑dataset comparisons. The model is lightweight (675k parameters, 2.6 MB footprint) with 1.5 ms/sample inference latency and demonstrates robust, interpretable forecasting with reduced drawdowns in algorithmic trading simulations.
The paper introduces iCReN, a framework for causal representation learning in time‑series data that jointly models instantaneous and lagged causal relations while handling nonstationarity. It establishes theoretical conditions for identifying latent states and their causal structures up to permutation and component‑wise invertible transformations, using an auxiliary variable linked to changes in transition‑noise distributions. Experiments show that iCReN accurately recovers latent states and both types of causal relations on synthetic data and improves downstream forecasting on real‑world data.
arXiv:2411. 17136v2 Announce Type: replace-cross Abstract: Realised volatility has become increasingly prominent in volatility forecasting due to its ability to capture intraday price fluctuations.
The paper introduces the Finance‑Aware Graph Spatio‑Temporal Network (FA‑GSTN) for forecasting realized volatility by treating the implied volatility surface as a dynamic graph. Nodes represent grid points on the surface, with edges capturing adaptive intra‑day spatial and explicit inter‑day temporal relationships, while finance‑aware node features (e.g., option Greeks) and a multi‑scale temporal smoothing gate address high‑frequency noise. Experiments on a large equity options dataset show FA‑GSTN achieves state‑of‑the‑art predictive accuracy (R² up to 0.473) and outperforms Vision Transformer baselines even with only one year of training data, demonstrating robustness during market stress.
arXiv:2610.10323v1 Announce Type: new Abstract: Short-horizon realized volatility forecasting requires the integration of market information that evolves at incompatible temporal resolutions, from se...
MoFE is a deep learning framework that combines Fourier Neural Operators with a Mixture-of-Experts architecture to forecast cryptocurrency prices. It models volatility as a mix of multi-frequency components—including fundamental growth, mining costs, halving events, and market sentiment—using adaptive FNO and convolutional experts. Experiments on Bitcoin data from 2020 to 2025 show MoFE outperforms existing models in short‑term horizons, reducing phase‑lag errors and improving directional accuracy and information coefficient, which translates into higher Sharpe ratios in simulated trading.
arXiv:2606. 09104v1 Announce Type: cross Abstract: Deep reinforcement learning (DRL) frameworks for portfolio optimization have shown promise for their ability to learn allocation rules dynamically from market data.
arXiv:2607. 22491v1 Announce Type: new Abstract: Volatility forecasting is dominated by persistence and measurement noise, leaving limited residual structure for nonlinear models to exploit.