New York Times: What the Past Can’t Tell Investors
I am linking to the story so as to create an opportunity for commenting, below.
New York Times: What the Past Can’t Tell Investors
I am linking to the story so as to create an opportunity for commenting, below.
You must be logged in to post a comment.
Finding one definition of momentum that doesn’t work doesn’t mean that momentum doesn’t work. If you switch to daily data and use a range of lookbacks with a range of markets, you will find that performance tends to improve over passive positions, even after transaction costs are accounted for.
A momentum variation that has worked is here:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=962461
Sustained, profitable momentum trading suggests that either information takes time to find its way into the market, or that traders are trading on past prices, i.e. “feedback trading”:
http://www.midasoracle.org/2009/11/30/robin-hanson-manipulations-prediction-markets/
Thanks.
Model now sees stocks heading south for the next two weeks; not least the resource stocks such as VED and XTA etc …
A couple of points: First, if the market is totally efficient, then alpha = 0. I think hindsight analysis pretty much shows that alpha > 0. Second, while momentum strategies tend to have higher volatility, they are also more leptokurtotic (think jLo), so the upside is higher than reversion strategies where you are catching the falling knives. My trading strategies are fairly agnostic along this line, but if I had to choose one, I would choose momentum (with special hedging) over reversion.
“Model now sees stocks heading south for the next two weeks; not least the resource stocks such as VED and XTA etc …”
And, what has happened – the market has headed south (down 400 points in two days) and the resource stocks have been hammered…….
So, what does this tell us?