A quant trading question I’ve never found a good answer to — how to decide between spending time improving an existing strategy (high certainty, low return on effort) vs starting something entirely new (uncertain but potentially high return on effort).
-
-
Replying to @macrocephalopod
I take a modular/alpha factory approach. A novel signal can usually be combined with existing risk, entry, execution etc. - if compute resources aren't a constraint you can test a lot of ideas continuously and quickly this way.
2 replies 0 retweets 18 likes -
Replying to @WallStreetCorgi
I do that for eg all strategies that fit within a framework (eg futures trading with up to a 1 minute rebalance). But what about stuff that fits outside the framework, eg should I start looking at equities, or crypto, or NDFs?
3 replies 0 retweets 7 likes -
Replying to @macrocephalopod @WallStreetCorgi
Is your example a matter of applying the same framework to a different universe, instead of an entirely different framework (i.e event-driven time series vs periodical rebalancing cross sectional stuffs)?
1 reply 0 retweets 2 likes -
I hope to be able to do the latter.
My hunch is having a different framework/approach can create more truly uncorrelated strategies, because there are only so many ways to slice and dice using an existing framework.1 reply 0 retweets 2 likes
Yeah I’m talking about the latter. Applying same framework to a new universe is generally (though not always) much lower effort.
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.