Online forecasting for financial time series
Nikolas Michael, Mihai Cucuringu, Sam Howison · 2026
Read Online forecasting for financial time series by Nikolas Michael, Mihai Cucuringu, Sam Howison (2026) 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/pmc-13675284.txt.
Description
Nikolas Michael, Mihai Cucuringu and Sam Howison introduce OFTER, an online forecasting pipeline for financial time series, in work published in 2026. The method targets mid-sized multivariate series, a common shape in finance. It pairs k-nearest neighbours and a generalized regression neural network with a dimension reduction step, using a weighted norm based on a tweaked correlation measure.
The design is built for online use, gives readable output and works in low signal-to-noise settings such as daily equity forecasting. Tests beat several leading baselines, the authors report. They argue that classic tools, carefully combined, can rival deep learning while adding scale and interpretability, which matter for live financial work.
Catalog record
| CREATOR | Nikolas Michael, Mihai Cucuringu, Sam Howison |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 167 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |