Nicholas Taleb, author of the bestsellers Fooled By Randomness and The Black Swan, on how we can protect ourselves against future black swans.
“What is fragile should break only while it is still small. Nothing should ever become too big to fail. Evolution in economic life helps those with the maximum amount of hidden risks – and hence the most fragile – become the biggest.”
Taleb doesn’t point out that there are drawbacks to robustness: you sacrifice performance in the present. However, as painful as it is not to be as efficient as you can today, I think he’s right – there’s no point in making a million dollars today if you lose it all tomorrow. The issue comes down to 2 choices: to specialize or to adapt. If we specialize, ie optimize our abilities, skills, and characteristics to our environment, we can do very well, but we’ll blow up when the environment changes. If we choose to be robust, we can only hope for average performance in any given environment, but we’ll survive almost any environmental change.
Bringing that issue to trading: I’ve chosen the latter route in designing my systems. My previous ATR systems performed very well – but they had 6 to 7 optimized parameters, so they were specialized to trade in that particular environment (perhaps too specialized). I’ve spent the last month working on a promising new system – the basic system only has 2 parameters, 4 if I include stops. It’s as simple as they come, and so far it’s worked on 5 markets that I’m interested in. Robustness is key – I tried to make the out-of-sample period as large as possible and checked to see if it performs well even if it’s optimized over a small period. I also tried to use markets where I’d have sufficient data to cover at least 2 different types of secular markets.
Performances were not spectacular, but I’m pretty confident that it’ll survive any sort of environment. Add that to appropriate money management and portfolio diversification, and I think I have a potentially viable trading system.