← Back to research

Nowcasting using regression on signatures

nowcasting
macroeconomics
signatures
modelling

Cohen, Samuel N., Giulia Mantoan, Lars Nesheim, Áureo de Paula, Arthur Turrell, and Lingyi Yang. Nowcasting using regression on signatures arXiv preprint arXiv:2305.10256v2 (2025).

Nowcasting UK unemployment
Authors
Affiliations

Samuel N. Cohen

University of Oxford

Giulia Mantoan

Bank of England

Lars Nesheim

University College London

Áureo de Paula

University College London

Arthur Turrell

Bank of England

Lingyi Yang

University of Oxford

Published

December 2025

Abstract

We introduce a new method of nowcasting using regression on path signatures. Path signatures capture the geometric properties of sequential data. Because signatures embed observations in continuous time, they naturally handle mixed frequencies and missing data. We prove theoretically, and with simulations, that regression on signatures subsumes the linear Kalman filter and retains desirable consistency properties. Nowcasting with signatures is more robust to disruptions in data series than previous methods, making it useful in stressed times (for example, during COVID-19). This approach is performant in nowcasting US GDP growth, and in nowcasting UK unemployment.

BibTeX citation

@misc{arxiv:2305.10256,
  author = {Cohen, Samuel N and Mantoan, Giulia and Nesheim, Lars and de Paula, Áureo and Turrell, Arthur and Yang, Lingyi},
  title={Nowcasting with signature methods},
  doi = {10.48550/arXiv.2305.10256},
  url = {https://arxiv.org/abs/2305.10256v2},
  publisher = {arXiv},
  year = {2025}
}