On the street, it generally means spread extraction, rebate trading, stat or index arb. I think the independent retail trader generally can't play spread extraction in liquid names, and rebate trading is out of the question. Stat arb is my retail approach, but how high-frequency I can get really depends on the data and the commission structure I get.
Are you implying your stat arb model is actually scalable in terms on holding period? Wouldn't the expected profit decrease with longer holding period since the model ideally works with higher frequency?
The holding time loosely scales with the time-frame I'm trading, yes. The profit potential doesn't decrease over a long period of time unless there's a structural break, and by structural break I'm referring to a formal definition. As a matter of practicality, there's no benefit to holding because structural breaks can and do happen, just on account of short-term news. I hope that makes sense.
just wondering how can a strategy that trades with a half-life of seconds be based on statistical arbitrage? Maybe we play with definitions here but I consider most short term market moves random and white noise unless you approach the whole story as a market maker, earning the spread. I am not familiar with market making in cash markets but more on the equity derivatives side. I know of some guys who trade high frequency strats prop based on flow and pick up in short term momentum but I would not define this as stat arb in the traditional sense. To me stat arb is the exploitation of divergences from the mean (however derived) beween two or more assets that generally highly correlate. Calling this even arbitrage is in my opinion an insult to highly paid financial mathematicians who often very successfully find most optimal hedges against exposure resulting from exotic derivatives that others are either unable to correctly price due to lack of understanding and technology. To comment on your last question I think its impossible to trade high frequency retail unless you are making markets and try to extract the spread, any other high frequency trading will eat you alive due to transaction costs. That last part is just my opinion I dont trade such strategies so you are much better off asking TSGannGalt or LoLatency among others.
How come most of you quant guys can't look @ charts @ trade with discretion to make money? It's not that hard. You all spend countless days & hours programming & focusing on the technology when you could be swinging 50-lots @ a time in ES & CL, etc.; do you not know how to trade? Or are you trying to create a program that will trade your style so you can sit on a beach somewhere?! ::Never understood this::
wow, are you serious? what do you think is easier to predict, what is going to happen in the 50 milliseconds or what is going to happen in the next 1 year??
absolutely in the next year, no second doubt about it!!! Unless you have access to order flow maybe you can point out where you take your prediction from what the next tick is gonna be and especially how you plan on capitalizing on such knowledge. By the way, I am not in the business of predicting, I trade AFTER catalysts hit the market, I am no crystal ball reader.
the difference is, at the 50ms end of the spectrum you will be able to more quickly generate a statistically signficant number of trades while being able to recalibrate and refine your model and adapt to changing market conditions... on the other end of the spectrum, not so much so... theoretical profits will also always be greater at the high frequency end of the spectrum due to the length of the coastline of the price path, but that is a different argument... good luck