Synthetic Volatility Index Quantified

In my post “Engineering a Synthetic Volatility Index” I discussed a technique that I’ve developed to monitor broad market volatility with a Synthetic Volatility Index. Today I’m going to quantify our index by applying it to a liquid universe of equities as a simple entry filter. If volatility indeed has an effect on price return then we should see high volatility environments, which are typically associated with bear markets, render lower returns when compared with low volatility environments, which are typically associated with bull markets.

To test this theory I first had to adapt our indicator to provide binary high/low volatility signals. I achieved this by taking a simple 40 period moving average of our synthetic volatility index. Periods characterised by rising volatility, or when our synthetic index is greater than its 40 period moving average, were classified as high volatility and vice versa. This is a simple and clear determinant of volatility that can be easily quantified.

I wanted to understand the effect of broad market volatility on individual equities. In particular, I wanted to get a sense of the effect broad market volatility has on short-term price returns, specifically five day returns. I quantified this by using our adapted binary high/low volatility indicator as the only filter or entry condition for individual stocks. For the exit I employed a simple five day time stop. Here are the precise rules that I used:

Entry condition High Volatility: Synthetic Volatility Index of JSE Top 40 Index > 40 period simple moving average of Synthetic Index

Entry condition Low Volatility: Synthetic Volatility Index of JSE Top 40 Index < 40 period simple moving average of Synthetic Index

Exit condition: Five day time stop

The results are in – Volatility has a significant effect on short-term price returns. In fact, our simple volatility filter boosts returns by a factor of 2 on the long side in low volatility environments. Furthermore, these findings are highly statistically significant: our sample includes over 300 000 trade samples. The results are also intuitive, adding to the robustness of our indicator even further. Here are the test results since 2003 across liquid listed and delisted JSE equities:

Average five day return high volatility: 0.17% and 146 134 trades in sample

Average five daye return low volatility: 0.34% and 183 501 trades in sample

The uses of our volatility indicator are manifold: One could employ our filter to move one’s portfolio to cash, reduce trade exposure, employ different strategies, close long positions or take a short bias. I’ve used variants of this very successfully through the years in my own trading and managing client funds.

For those that use Metastock you can easily apply the indicator to your strategies or the charts with the following code:

High Volt: Security( “Path to JSE Top 40 Index”, MOV(ATR(1) / Close, 20, S) > MOV(MOV(ATR(1) / Close, 20, S) , 40, S) )

Low Volt: Security( “Path to JSE Top 40 Index”, MOV(ATR(1) / Close, 20, S) < MOV(MOV(ATR(1) / Close, 20, S) , 40, S) )

We welcome your thoughts and comments.

Happy Trading,
PJ

Passionate about generating and sharing quantified trading models that empower individuals to trade successfully. I founded www.sutherlandresearch.com to realise my passion.

1 comments On Synthetic Volatility Index Quantified

Leave a reply:

Site Footer