For years we at Huygens have been using and improving our stock market volatility prediction systems to drive our tactical investment strategies.

Our tactical strategies are designed to capture gains in periods of low and declining volatility, and to rotate to defensive portfolios when our systems predict that stock market volatility is increasing or abnormally high.

Why predict stock market volatility? Two reasons.

First: Higher stock market volatility is strongly linked to likelihood of investment losses.

Consider the following analysis, which breaks the timeframe 1992 through 2018 into 5-day periods. The grey line shows the familiar S&P 500 index cumulative price return, with the 5-day periods ordered sequentially by date.

The blue line shows the cumulative return when those same 5-day periods are re-ordered by volatility. The least volatile periods are on the left side of the chart, and the most volatile periods are on the right side. You can see that all of the S&P 500’s gains, and then some, were earned in the least volatile periods. Much of those gains were given up in the most volatile periods.

So if it’s possible to predict volatility with reasonable certainty (and it is), then it should be possible to avoid some of the largest losses.

SP500 returns and vol.gif

 

Second: Certain information can be used to predict equity market volatility.

Consider for example the persistence of volatility. The below chart uses the same 5-day periods as mentioned above, but this time plots the daily return volatility of period t (on the x axis) against the daily volatility of period t+1 (on the y axis).

The positive trend line slope and R-squared of +0.42 demonstrate that even something as simple as last week’s volatility can be useful to predict future market volatility, and therefore the likelihood of future losses.

Volatility persistence.gif

This is just an illustrative example. Our volatility prediction signals use more information than this to predict stock market volatility and the likelihood of large losses.


Our volatility prediction signals

Our proprietary volatility prediction signals for our tactical equity and tactical derivatives strategies update daily and will typically switch between portfolios 10-20 times per year. We use mainly S&P 500 derivative instrument pricing as input, as well as other stock market metrics, to look for 4 key conditions in investor sentiment:

  • Strongly bullish sentiment

  • Extremely bearish sentiment

  • Sentiment trending more bullish

  • Sentiment trending more bearish