Answers are grounded in this paper and cite their sources.
This paper introduces a robust and easy-to-implement one-pass market-beta estimator. It only requires first winsorizing daily stock rates of return at −2 and +4 times the contemporaneous market rate of return. In predicting future market-betas, this “slope-winsorized†beta estimator predicts future betas better not only than OLS betas, Bloomberg betas (ubiquitous on financial websites), and Vasicek (1973) betas, but also published estimators that require intra-day data, super-computers, or financial statements. Moreover, using weighted-least squares to exponentially decay the weight of aged return observations (with a half-life of about four months) further improves the estimates.
No takes yet. Share an insight, caveat, or question.
Ivo Welch (2022) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: