Breadth Volatility and Stock Market Cycles
Above is a measure that I've been working with lately that I call breadth volatility. The indicator is created by tracking the percentage of stocks trading above various short term moving averages and then calculating the standard deviation of that series. What we find is that breadth volatility tends to be low ahead of cycle peaks and crests as markets make--and rally from--bottoms. Going back to 2012, when breadth volatility has been in its highest quartile (greatest volatility of breadth), the next five days in SPY have averaged a gain of +.54%, much greater than the +.10% for all other occasions in the sample.
Interestingly, the breadth volatility measure shown above correlates only .41 with VIX going back to 2012. That's a significant correlation to be sure, but suggests that VIX only accounts for 16% of the variance in breadth volatility. Indeed, when I've tracked breadth volatility for specific VIX regimes, the predictive results with SPY have been quite good.
As you can see from the chart above, we made a low in breadth volatility ahead of the recent price peak in stocks and now have been trending higher as the market has sold off. As of yesterday's close, we were not yet at levels associated with intermediate-term market bottoms.
Further Reading: Perspectives on Stock Market Breadth
.
Interestingly, the breadth volatility measure shown above correlates only .41 with VIX going back to 2012. That's a significant correlation to be sure, but suggests that VIX only accounts for 16% of the variance in breadth volatility. Indeed, when I've tracked breadth volatility for specific VIX regimes, the predictive results with SPY have been quite good.
As you can see from the chart above, we made a low in breadth volatility ahead of the recent price peak in stocks and now have been trending higher as the market has sold off. As of yesterday's close, we were not yet at levels associated with intermediate-term market bottoms.
Further Reading: Perspectives on Stock Market Breadth
.
Comments
Post a Comment