Turning points are the Achilles' heel of time-series momentum portfolios. Slow signals fail to react quickly to changes in trend while fast signals are often false alarms. We analyze how momentum portfolios of various intermediate speeds, formed by blending slow and fast strategies, handle turning points. We find that the intersection of slow and fast signal directions possesses predictive information, including predictably negative returns when both signals are negative. We propose a novel decomposition of momentum strategy alpha, highlighting the role of volatility timing; and a mean-variance optimal dynamic speed-selection strategy with efficient out-of-sample performance across international equity markets.

Keywords: time-series momentum, volatility timing, market timing, asset pricing, trend following, turning points, momentum speed, mean reversion, behavioral finance

Continue reading here.