A General Class of Model-Free Dense Precision Matrix Estimators
Mehmet Caner, Agostino Capponi, Mihailo Stojnic · 2025
Read A General Class of Model-Free Dense Precision Matrix Estimators by Mehmet Caner, Agostino Capponi, Mihailo Stojnic (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-2507-04663.txt.
Introduction
Mehmet Caner and co-authors work on a basic tool of statistics, the precision matrix. This matrix sits at the heart of work in economics and finance. Good estimates of it should be dense, steady, and free of strict model demands. Past methods could hit one or two of these goals, but not all three at once.
The team builds a broad class of estimators that hits all three within one frame. Their study pins down how error turns on three things: the cost of the method, the strength of the signal, and bias. One plain member of the class shows double descent, a quirk known from regression. Tests on a major stock index show a matching pattern in returns.
Catalog record
| CREATOR | Mehmet Caner, Agostino Capponi, Mihailo Stojnic |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 166 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-04663