Short thread about momentum as a risk factor and why you should take it seriously. At first glance it sounds stupid — why would stocks that recently went up continue to move together in the future?
Show this thread
-
It’s common to neutralise momentum to other factors (eg sectors, countries, size etc). In the limit where you neutralize to all known factors, momentum is a summary of exposure to all the *unknown* factors, the stuff you didn’t model because you didn’t know about it.
Show this thread
That’s why you should include momentum in your risk model even if you don’t plan to use it as a source of alpha — it’s a simple, robust way of modelling the risks that you aren’t smart enough to think of in advance.
-
-
Never thought about it this way--momentum just being a constant way of factoring thematic baskets that haven't been factored out yet. Super interesting, thanks.
-
It gets even more interesting when looking at it in a granular basis through specific names that one might consider more closely related than the market reflects and how they eventually converge into a stronger correlation
End of conversation
New conversation -
-
-
do you have any papers you would recommend for thinking about this more? super interesting concept!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Estimate errors.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Great thread - also momentum factor does contain significant left tail risk ('mom crash') which is arguably part of the momentum premium
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
A great take on MOM, I'd never seen it cast in this light. If folks are looking for a good academic treatment, the Daniel et al. Momentum Crashes paper is a nice start.https://www.sciencedirect.com/science/article/pii/S0304405X16301490 …
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
It almost sounds like you're talking about using synchronization of stock movement as a proxy for unknown factors at play. Isn't this different than momentum? What am I missing?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
The LSE did a study on this 20-30 years ago. random groups of traders (students I believe) were given money to play with and after a year all had roughly the same returns. But half of those had an enforced momentum trading rule, they had better risk-adjusted returns
-
You can do a cute experiment with historical data where you enter random trades and then enforce a rule which exits the losers (and replaces them with more random trades) and holds on to the winners. It makes money on average.
End of conversation
New conversation -
-
-
I think there is a gem of an idea in here, but I think you need to bring it down almost to equations to really be able to see if what you are saying makes sense.
-
Agree equations are better, Twitter doesn't lend itself very well to that unfortunately.
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.