Unsupervised ML clustering of assets into equally weighted sub-portfolios dominates both plug-in minimum variance and equally weighted benchmarks, with optimal cluster count around …
Correcting for survivorship and delisting bias substantially overturns published cryptocurrency anomaly evidence: size is overstated, while momentum and market beta no longer price …
In this paper, we develop a novel, intuitive and objective measure of time-varying parameter uncertainty (PU) based on a simple statistical test.
We show that parameter uncertainty based on the turbulence within each cross-section of factor portfolios produces a significant out-of-sample forecast for six out of seven tested …
Based on a multidisciplinary literature review, we discuss the assumptions implicit in the prevalent Black-Scholes model and argue for relaxed assumptions that better represent …
Do forecasting errors from direct predictions of market components out-way the additional errors introduced by an intermediary asset pricing model?