{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T18:00:41Z","timestamp":1791396041743,"version":"4.3.3"},"posted":{"date-parts":[[2026]]},"group-title":"SSRN","reference-count":0,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>Individual stock-return forecasting is inherently a low signal-to-noise problem. This reality has led to forecast evaluations in which the zero forecast is routinely used as a benchmark. We document that this benchmark is excessively conservative. Outperforming the zero forecast, in fact, may simply reflect ability to predict level effects, i.e., changes in the market return or market-timing, rather than cross-sectional dispersion. Consistent with this observation, we document that the panel historical mean provides a more stringent and informative benchmark. The panel historical mean is economically interpretable as the historical mean return of the market portfolio and is nested within standard machine-learning models as the limiting case in which firm characteristics have no predictive role. We formalize the difficulty of outperforming this new benchmark through a &lt;i&gt;finite-sample&lt;\/i&gt; analysis of Ridge in a &lt;i&gt;low &lt;\/i&gt;signal-to-noise environment. Relative to the panel historical mean, only flexible nonlinear models generate consistent predictive gains, which are - however - concentrated among small, value, and illiquid stocks.<\/jats:p>","DOI":"10.2139\/ssrn.7573698","type":"posted-content","created":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T17:38:04Z","timestamp":1791394684000},"source":"Crossref","is-referenced-by-count":0,"title":["Noisy Financial Machine Learning"],"prefix":"10.2139","author":[{"given":"Federico Maria","family":"Bandi","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sofonias Alemu","family":"Korsaye","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yinan","family":"Su","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","container-title":[],"original-title":[],"deposited":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T17:38:04Z","timestamp":1791394684000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ssrn.com\/abstract=7573698"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":0,"URL":"https:\/\/doi.org\/10.2139\/ssrn.7573698","relation":{},"subject":[],"published":{"date-parts":[[2026]]},"subtype":"preprint"}}