The Time-partitioning Gaussian Process Mixture Model (TPGPMM) is a Bayesian nonparametric framework that identifies persistent regimes and models nonstationary dependence in time series. It uses an adaptive Pitman–Yor prior on contiguous segmentations combined with regime-specific Gaussian processes, and learns change points and hyperparameters via a split–merge Metropolis-within-Gibbs sampler. Experiments on the CBOE Volatility Index and the 10‑year U.S. Treasury yield recover regime structures aligned with major market and monetary‑policy events, while variants that estimate observation noise improve predictive accuracy but may reduce sparsity or event alignment.
arXiv:2606. 03184v1 Announce Type: cross Abstract: Financial forecasting is difficult due to low signal-to-noise ratios, latent factors, heavy tails, regime shifts, and jumps.
arXiv:2606. 02657v1 Announce Type: new Abstract: The standard generalization bounds assume that the training and deployment distributions are the same, or are static, and don't consider regime switching environments where the ratio of calm vs crisis states is different.
The paper introduces latent jump models (LJMs) for continuous-time latent state-space modeling, where the latent state evolves deterministically between Poisson-driven events and experiences learned instantaneous jumps at each event. It presents a tractable event-augmented ELBO and a novel Scan-Parallel Variational Inference (SPVI) method that samples the full latent event configuration before evaluating state recurrence, enabling logarithmic temporal depth and fast, scalable training. Experiments on synthetic and real irregular time series show that LJMs achieve competitive forecast accuracy with fewer latent updates and train one to two orders of magnitude faster than sequential baselines.
arXiv:2605. 12764v3 Announce Type: replace-cross Abstract: This paper introduces a physics-informed generative framework that resolves the fundamental conflict between the statistical flexibility of deep learning and the rigorous theoretical constraints of fixed-income modeling.
arXiv:2605. 30363v2 Announce Type: replace-cross Abstract: Regime shifts in financial markets reorganise the joint dynamics of asset prices and macro variables, breaking any single-regime calibration.
arXiv:2605.23632v2 Announce Type: replace Abstract: We introduce Gaussian Mixture Copula Processes (GMCP), a conditional copula process for irregularly sampled multivariate time series (IMTS) that is...
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:2603. 15802v2 Announce Type: replace Abstract: In many time series forecasting settings, the target time series is accompanied by exogenous covariates, such as promotions and prices in retail demand; temperature in energy load; calendar and holiday indicators for traffic or sales; and grid load or fuel costs in electricity pricing.
arXiv:2608.23802v1 Announce Type: cross Abstract: Many common data dependencies can be characterized by graphs: time series data are sequential (chain graph), images appear as pixels (lattice graph),...
arXiv:2606. 14313v1 Announce Type: cross Abstract: Real-world spatio-temporal forecasting must handle irregular time points, spatially sparse observations, and the need for uncertainty quantification.
arXiv:2609.37715v1 Announce Type: new Abstract: Time-series foundation models are increasingly adapted to new domains through fine-tuning on target data, under the implicit assumption that more targe...
arXiv:2608. 13621v1 Announce Type: new Abstract: A hidden Markov model (HMM) combines three roles: inference of a hidden-state belief from observations, propagation through a Markov transition, and emission back to observation space.