High Frequency Trading - Hype or Substance?

Discussion in 'Strategy Building' started by CPTrader, Jul 6, 2005.

  1. MacroEvent still around? This vol is a b*tch or a blessing depending on the design for small martingale strategies.
     
    #111     Mar 31, 2009
  2. TeganZ

    TeganZ

    Here is a great resource on HFT:

    High Frequency Trading

    Kate
    Advocate of CBOE
     
    #112     Mar 15, 2010
  3. ET580

    ET580

    dersu:

    Recently I've been noticing problems with the quality of my datafeed, which I want to improve, while looking into some high-frequency trade techniques so I need faster as well as better.

    What is your solution for high-quality (minimum latency) data? And is it available for others?
     
    #113     Mar 18, 2010
  4. Thanks, will go through it.

    More please...
     
    #114     Mar 18, 2010
  5. Cambist

    Cambist

    If any of you truly want a Q&A regarding HFT you can feel free to PM me. I worked for a couple years as the FX desk manager at a HFT hedge fun in Chicago. The concept is actually quite simple and transaction costs not an issue. We accounted for 15-20% of CME volume in most of the majors and raked in about 5-10x the average american salary per day on our 8 person desk (firm was 250 people and a desk for any asset class).

    The barrier of entry is the huge cost of technology.
     
    #115     Mar 18, 2010

  6. I also have fun hedging around....
     
    #116     Mar 18, 2010
  7. What sort of HFT strategies were you running? Arb based?

    Am guessing round trip messaging times were in the 100ms bracket. What were the position holding times? We talking ms ,sec, min?

    You say you traded all the majors. Didn you find the small contract size on the 6B troublesome from a comms perspective?

    Personally I think it's ridiculous the Merc lump this contract into the same fee bracket as the bigger FX contracts such as the 6E.
     
    #117     Mar 18, 2010
  8. I doubt these are arbs, I'm thinking that they are all information gathering to obtain information about liquidity, market participant stops, etc.. you plug that into a more run of the mill strategy and then crank it up a few notches, it should work for a couple of years.
     
    #118     Mar 18, 2010
  9. You couldn't be more wrong. The problem is that you don't have the knowledge necessary to correctly evaluate HFT strats. For example, it's clear from your post that you are not aware of how extremely high-frequency trading systems are set up to execute. Here's a hint for you. Do you buy AAPL at 220.0187? The answer would be no. Can you execute at 220.0187? Again, no.

    This is not meant to be a putdown. Just meant to let you know that you don't (and, most likely, couldn't possibly) have the type of knowledge necessary to evaluate these strats because the type of trading involved is something you've never considered. It's hard to get the information you need, but it's possible. If you have very good programming skills, maybe you should seek some of it out.
    Manual execution off of 1 minute bars is not HFT as it is being discussed here. HFT is thousands running up to tens of thousands of trades a day executed automatically.
     
    #119     Mar 18, 2010
  10. Cambist

    Cambist

    This guy is probably the only other person in here that knows what he's talking about.

    I offered to chat with people via IM and received a bunch of them. I'll be trying to respond in the next couple days. But what it comes down to is that HFT is nothing what everyone here believes it to be.
     
    #120     Mar 24, 2010