Can A Volatility Trsunami Be Predicted?

In algorithmic trading and risk management system development, a quest for the “holy grail” has been centered on creating a system that warns of forthcoming market volatility. While no system of risk management is perfect at warning about a significant stock market decline, Andrew Thrasher, a portfolio manager at Financial Enhancement Group, recently was the 2017 Charles Dow Award winner for a paper he wrote on “Forecasting a Volatility Tsunami.” In the paper, Andrew Thrasher makes the assertion that it is the suppression of volatility returns dispersion that has, in the past, led to higher market volatility.

Andrew Thrasher – Missing the top ten performing stock market days hurts, but avoiding the top 10 losing days helps

There is a well-worn statistic in investing that missing the ten best days of the stock market can significantly reduce returns, but there is a little-discussed flipside to that coin.

“Those that believe in the adage of buy-and-hold investing often mention that missing the ten or twenty best trading days has a substantially negative impact on a portfolio’s overall return,” Thrasher wrote. “They then, in turn, reject the idea of attempting to avoid the worst days in the market and active management as a whole.” In other words, they provide credence to upside volatility (deviation) but don’t give the same level of respect to downside volatility.

Looking at the stock market bull run from 1984 to 1998, for instance, researcher Paul Gire found that if an investor was out of the market on the top ten performing days, annualized returns dropped more than 3% from 17.89% to 14.24%, a statistic cited by passive investing advocates to justify a buy and hold strategy.

What is mostly left unsaid, Thrasher notes, is that missing the ten worst stock market declines over that period would have boosted annualized returns by over 6%, from 17.89% to 24.17%. In fact, if one were to miss both the ten worst and ten best stock market sessions returns would jump to 20.31% with significantly lower volatility. In his analysis, Thrasher also notes that periods of market high and low points occur in a tight pattern. Nearly half of the worst and best-performing stock market days were no more than 12 days apart.

The issue is how can volatility, which by definition is a market “surprise,” be forecast?

Andrew Thrasher – Suppression of returns dispersion has led to volatility

There is a belief among some market participants that when volatility moves to abnormally low levels this is a warning sign that a surprise market event might be on the horizon.

Thrasher, however, believes the concept is not an absolute low number on the VIX that foretells potential market danger, but rather a pattern of suppressed returns. “There is a more optimal tsunami-like condition that takes place within the markets, providing a better indication of potential future equity market loss and Volatility Index increase,” he wrote.

What Thrasher watches is the dispersion of returns. Using a 20-day standard deviation measure, he notes that when the index standard deviation over that period is low it has a tendency to be followed by a larger than average volatility spike. Thrasher notes many of the spikes that occurred over the last ten years have been preceded by very narrow dispersion within the Volatility Index. Said differently, while not all narrow trading ranges within the VIX are followed by spikes, nearly all spikes in volatility have followed tight trading ranges.

“It has been shown that the evaluation of the dispersion within the VIX and VVIX act as accurate barometers for future large advances in the Index,” he wrote. “The majority of spikes that have taken place in the Index occur after the dispersion of the VIX has fallen below the specified threshold (by a standard deviation.”

In order to find an appropriate threshold with forecasting spikes in the Volatility Index, the daily standard deviation readings were ranked by percentile for the time period of May 2006 through June 2016. As a result, the fifteenth percentile allowed a sizable sample size of 373 to be obtained. The fifteenth percentile standard deviation during the above-mentioned timeframe for the Volatility Index is 0.86. Chart 5 shows the scatter plot of the data observed for the 20-day standard deviation for the VIX and the resulting three-week maximum change in the Index, which was calculated by using the highest high in the subsequent fifteen trading days for each data point. By looking at the maximum change in the VIX we can begin to see that the largest spikes within a three-week period occur when price dispersion is extremely low; while the three week maximum change in the VIX diminishes the larger the dispersion becomes.

When volatility is in a very tight, congested trading range, the average volatility spike following this signal has been up on average 34.3% compared to a normalized rise in volatility of 14.95%, according to Thrasher.

Like all methods, the indicator is not perfect. “Similar to the suboptimal method of using large declines in the VIX as a predictor of future spikes, the VVIX dispersion threshold has many false signals that are now followed by volatility spikes,” he writes.

For his part, Thrasher uses volatility dispersion to recognize forthcoming market environments and manage risk in equity portfolios. The signal is used in combination with other technical indicators, such as option flow, put call ratios, measures of breadth, shorter intraday price movements, as well as backwardation and contango in the VIX futures curve.

Andrew Thrasher