Work

A Jump-Diffusion Framework for Irregular Time Series Generation

Year: 2025

Type: preprint

Abstract: We propose a framework for generative modeling of continuous-time processes from irregularly and asynchronously recorded data. It is based on the matching of generators and accommodates discontinuous trajectories. Analytical formulas for diffusion and jump bridges yield a family of reference generators that a neural network is trained to match. The key ingredient is that, for our constructed jump bridge, a parametrization of the jump kernel densities by scaled Gaussians admits closed-form expressions for the Kullback-Leibler divergence, allowing simulation-free training.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2505.23215

Authors: Pfohl, O., Chemseddine, J., Hagemann, P., Steidl, G., Wald, C., Jahn, T.

Institution: Technische Universität Berlin

Language: English

Topic: Stochastic processes and financial applications

Subfield: Finance

Field: Economics, Econometrics and Finance

Domain: Social Sciences

Open Access status: green

Funders: Deutsche Forschungsgemeinschaft, Berlin Mathematics Research Center MATH+

Awards: G1372115314, G2033548659, G2585135387, G442688124, G5853193109, EXC 2046: MATH+: Berlin Mathematics Research Center, Bewegungsdetektion in der optischen Diffraktionstomographie und der dynamischen Einzelmolekülmikropkopie, G8964763783

OpenAlex ID: W4416610240

Cites:

Cited by:

Related to: