Conditional Neural SDEs for Financial Time-Series Generation
Abstract
Conditional generation of time-series paths is a central problem in probabilistic modeling, where the goal is to simulate future trajectories that are both realistic and responsive to external context. It has direct applications in finance and economics. Neural stochastic differential equations provide a natural framework for this task by combining continuous-time stochastic dynamics with learnable drift and diffusion functions. However, directly modeling the target state can be unstable in conditional settings, especially when the target and conditioning variables have strong level dependence, limited training data, or different dimensions.We propose a residualanchor Neural SDE architecture that decomposes the target into a learned conditional anchor and a stochastic residual. The anchor captures the context-dependent baseline, while the Neural SDE (Stochastic Differential Equation) models the residual dynamics around this baseline. This separation reduces the burden on the stochastic dynamics and generalizes spread-based representations to more general conditioning settings.We evaluate the approach on real-world financial time series, including conditional interest-rate scenarios and a comparison with a time-series foundation model. The results show that the proposed model can generate plausible conditional paths, maintain competitive point accuracy, and provide calibrated uncertainty estimates.
Related
- HyperNSDE: Personalized Neural SDEs for Joint Static-Longitudinal Clinical Data Generation
- Flowz: When Does Learning Forecast Errors Add Value? Conditional Flow Matching for Financial Price Forecasts
- SDECast: Probabilistic Weather Forecasting in Continuous Time with Neural SDEs
- What drives generation Z's appetite for crypto investment? The role of financial literacy, gender, and channels of influence
- Structured Neural SDEs for Functional Calibration
BibTeX
@misc{molavipour2026conditional,
title = {{Conditional Neural SDEs for Financial Time-Series Generation}},
author = {Sina Molavipour and Alireza M. Javid and Björn Löfdahl and Mikhail Nechaev},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.41405},
eprint = {2610.41405},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.41405},
}
RIS
TY - UNPB
TI - Conditional Neural SDEs for Financial Time-Series Generation
AU - Sina Molavipour
AU - Alireza M. Javid
AU - Björn Löfdahl
AU - Mikhail Nechaev
PY - 2026
DA - 2026/10/30
PB - arXiv
DO - 10.48550/arxiv.2610.41405
UR - https://arxiv.org/abs/2610.41405
AB - Conditional generation of time-series paths is a central problem in probabilistic modeling, where the goal is to simulate future trajectories that are both realistic and responsive to external context. It has direct applications in finance and economics. Neural stochastic differential equations provide a natural framework for this task by combining continuous-time stochastic dynamics with learnable drift and diffusion functions. However, directly modeling the target state can be unstable in conditional settings, especially when the target and conditioning variables have strong level dependence, limited training data, or different dimensions.We propose a residualanchor Neural SDE architecture that decomposes the target into a learned conditional anchor and a stochastic residual. The anchor captures the context-dependent baseline, while the Neural SDE (Stochastic Differential Equation) models the residual dynamics around this baseline. This separation reduces the burden on the stochastic dynamics and generalizes spread-based representations to more general conditioning settings.We evaluate the approach on real-world financial time series, including conditional interest-rate scenarios and a comparison with a time-series foundation model. The results show that the proposed model can generate plausible conditional paths, maintain competitive point accuracy, and provide calibrated uncertainty estimates.
ER -