Quote from Algo_Design_Kid:
Let me save you the time. I would shy away from the fantasy of stat arb. The best definition really means free money on a very short term time horizon.
There are 2 problems for you in regards to this subject.
1. There are too many players in this game already.
2. They all have co-location and can work these opportunities well before you. This is assuming you do not have the capability.
Another problem is that you might "see" this opportunity, but when it comes down to the real trading environment and not a live simulator these opportunities do not appear anymore. You will get eaten by commissions and the spread.
By all means try it out though, maybe you excel in something that others do not. I just wouldn't put all your effort into this subject because there is a very high probability that you will be let down.
If this indeed does fail just remember not to give up. There are plenty of ways to earn a few ticks here and there.
I completely agree with you in regards to the true definition of arbitrage. But what I am referring to and what I think most people mean when they use the term, is any strategy that uses statistical methods to generate trading signals.
As for co-location, I am not going to be focusing on hfT cause I don't believe there is an edge there for me. I am working on ideas that basically gets signals on daily prices that I am then able to use on a intra-day time frame whether its pairs trading, ETF vs components, mean reversion.
I understand that I am not able to compete with the hedge funds and prop desk that co-location servers and hundreds of millions of dollars, but I think that me being small with a few hundred thousand in buying power that I able to develop strategies that a lot of these guys over look.
But like you said I am going to give it a try and keep going till I find something.
