Neural MJD: Neural Non-Stationary Merton Jump Diffusion for Time Series Prediction

Yuanpei Gao [email protected], Qi Yan [email protected], Yan Leng [email protected], Renjie Liao [email protected] · 2025

Read Neural MJD: Neural Non-Stationary Merton Jump Diffusion for Time Series Prediction by Yuanpei Gao [email protected], Qi Yan [email protected], Yan Leng [email protected], Renjie Liao [email protected] (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-04542.txt.

Description

Gao and co-authors build Neural MJD in 2025, a forecasting model that writes prediction as a drifting, jumping random process with a neural brain. Plain deep models forecast well but hide their logic and stumble when data shift or jump. Their model blends a slow drift, random jitter, and sudden leaps, all shifting over time.

A cap on jumps per short span keeps the math workable, with a proven error bound. A restart solver cuts error and spread against the standard solver. On fake and real series, the model beat top deep and classic rivals. The design suits finance and other jumpy data.

Catalog record

CREATORYuanpei Gao [email protected], Qi Yan [email protected], Yan Leng [email protected], Renjie Liao [email protected]
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE78 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2506-04542

Browse the full library · About the Tower of Babel