{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T06:56:57Z","timestamp":1790578617582,"version":"4.1.0"},"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>Volatility and correlation need not benefit from the same amount of historical information. We treat their history lengths as separate empirical quantities and measure the cost of forcing them to coincide. Using daily U.S. equity returns, we first hold information quantity fixed while moving observations further into the past. An equally sized historical block becomes poorly representative of current volatility sooner than it becomes poorly representative of current correlation. We then repeat the same correlation-history comparison as the number of assets rises through fifty nested security sequences. The benefit of deeper correlation history increases at every dimension step, while the penalties associated with long volatility histories remain broadly unchanged. Across two out-of-sample evaluation periods using the same fixed security population, estimators constrained to one history length choose substantially longer histories in the first period. The difference coincides with a much larger cost from using long volatility histories in the second period, even though deeper correlation history remains highly valuable there. A pair of history lengths chosen from the first period continues to improve covariance forecasts when carried into the second, including against a one-decay EWMA selected using the later evaluations. Separating aggregate volatility magnitude from relative cross-sectional dispersion shows that the preference for recent volatility information is not confined to the market-wide level. In the second evaluation period, separate histories do not lower realised global minimum-variance portfolio variance, while deeper correlation history reduces turnover increasingly as the number of assets grows. A common covariance history is therefore a compromise between estimation problems with different responses to information age and dimension, rather than a measure of market memory.<\/jats:p>","DOI":"10.2139\/ssrn.7525858","type":"posted-content","created":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T06:29:56Z","timestamp":1790576996000},"source":"Crossref","is-referenced-by-count":0,"title":["When One Window Fits None: Volatility Recency, Correlation Depth, and the Cost of a Common Covariance History"],"prefix":"10.2139","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4502-3818","authenticated-orcid":true,"given":"William","family":"Smyth","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","container-title":[],"original-title":[],"deposited":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T06:29:56Z","timestamp":1790576996000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ssrn.com\/abstract=7525858"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":0,"URL":"https:\/\/doi.org\/10.2139\/ssrn.7525858","relation":{},"subject":[],"published":{"date-parts":[[2026]]},"subtype":"preprint"}}