Insteresting comment from Jason Ruspini.
Jason, the high-end event derivative traders at BetFair (and probably at TradeSports and Betdaq) do benefit from volatility —-they would trade thousands of times on one prediction market, back and forth, taking advantage of small price moves. Any related thought about that, with respect to your comment at Potfolio?
–
Previous blog posts by Chris F. Masse:
- “Is Clinton’s Pennsylvania Lead Really 20 Points?”
- The Most Surprising Piece Of News I’ve Heard Today
- My first prediction market plugin for WordPress
- Self-Serving Prediction Market Of The Day — Unlawful Internet Gambling Enforcement Act of 2006
- Prediction markets tend to be so illiquid, though, that mere activity looks like volatility.
It depends what you mean exactly by “benefit by volatility”.
.
First, managed futures strategies typically thrive on momentum, not mean reversion. In terms of volatility, technically they are long something more like gamma than vega. That is, more long trendiness than choppiness. The reason they do relatively well in high vol environments is mostly due to those funds 1) being relatively nimble in terms of putting on short trades in risky assets (while being long those assets in other periods) and 2) trading a wider range of markets. Many of the commodities these funds trade are relatively uncorrelated to traditional assets and relatively insensitive to risk appetites. This is also the reason why managed futures is the most difficult hedge fund style to model via factor analysis. If you are interested in that sort of thing, see this..
http://www.allaboutalpha.com/b…..-practice/
.
What you describe is a mean reversion strategy that would seem to rely more on having some minimum level of volatility that doesn’t vary too much than actually being long volatility. In other words, mean reversion strategies don’t generally get better when volatility increases – because direction (momentum strategy profits) often comes alongside increased volatility *in practice*. So I don’t quite agree with your identification of trading a range and benefiting from volatility.
.
Of course, if a market is liquid/active enough and the fees are low enough, market-makers can be profitable by playing the bid-ask spread. Prediction markets tend to be so illiquid though that mere activity looks like volatility. In terms of algorithmic trading of prediction markets, I tend to think that momentum strategies would work better on average, if only because binary contracts involve a drift towards zero, and such strategies would involve fewer trades and thus less slippage and fees. I have never run any of these tests though on prediction markets.
.
A partially related question is whether the volatility of prediction markets can be used as a confidence band. These would be good undergrad projects.
It depends what you mean exactly by “benefit by volatility”.
.
First, managed futures strategies typically thrive on momentum, not mean reversion. In terms of volatility, technically they are long something more like gamma than vega. That is, more long trendiness than choppiness. The reason they do relatively well in high vol environments is mostly due to those funds 1) being relatively nimble in terms of putting on short trades in risky assets (while being long those assets in other periods) and 2) trading a wider range of markets. Many of the commodities these funds trade are relatively uncorrelated to traditional assets and relatively insensitive to risk appetites. This is also the reason why managed futures is the most difficult hedge fund style to model via factor analysis. If you are interested in that sort of thing, see this..
http://www.allaboutalpha.com/b…..-practice/
.
What you describe is a mean reversion strategy that would seem to rely more on having some minimum level of volatility that doesn’t vary too much than actually being long volatility. In other words, mean reversion strategies don’t generally get better when volatility increases – because direction (momentum strategy profits) often comes alongside increased volatility *in practice*. So I don’t quite agree with your identification of trading a range and benefiting from volatility.
.
Of course, if a market is liquid/active enough and the fees are low enough, market-makers can be profitable by playing the bid-ask spread. Prediction markets tend to be so illiquid though that mere activity looks like volatility. In terms of algorithmic trading of prediction markets, I tend to think that momentum strategies would work better on average, if only because binary contracts involve a drift towards zero, and such strategies would involve fewer trades and thus less slippage and fees. I have never run any of these tests though on prediction markets.
.
A partially related question is whether the volatility of prediction markets can be used as a confidence band. These would be good undergrad projects.