Trying to make it on my own

Discussion in 'Journals' started by doublet83, Jul 1, 2011.

  1. Went live on a new quant based pair trading strategy today. I originally wanted to learn C# to automate this thing, but decided that it would be too much work for the time being. Thus, I did some ghetto backtesting in excel, using a bunch of if statements.

    Strat seems to work okay in the backtest, with 210% returns across 162 trades during a 9 month period. Average returns of around 1.3% a trade with a 15 day holding period and a profit factor around 2.0. Seems like okay results for an out of sample backtest. This does not include comissions or slippage, which I think may run me 30bps. However, the strat assumes a fill at the close, and holding period of a constant 15 days. Hopefully the fact that I have more flexibility in actual trading will improve results somewhat.

    I'm very curious what kind of returns others have achieved with pairs trading strategies. I believe my model is doing something unique here, although at times I wonder if all those calculations are adding any value.
     
    #21     Jul 12, 2011
  2. cloudy

    cloudy

    #22     Jul 12, 2011
  3. Damn, I took an nap today and now I can't fall asleep! The freedom of taking naps during the day has its downsides..

    Thanks for the compliment, although real results can different greatly from the backtest. I have some confidence that at a minimum, this strategy will not be a money loser. So I am willing to move quickly to live trading without that much backtesting.
     
    #23     Jul 12, 2011
  4. You've left it too late.
     
    #24     Jul 12, 2011
  5. Always exciting to have a new strategy up and running. Just hope this doesn't distract me from my core trading strategy.

    MTD performance by strategy:

    Core trading strat +3k
    Pairs trading strat +1.5k
    Long term strat +11k

    YTD performance by strategy

    Core +228k
    Pairs +1.5k
    LT +39k
     
    #25     Jul 14, 2011
  6. Good day today, with +13k of profits in the core strategy.

    However, was up 25k at one point, so wish I could have taken some profits, although I don't really like taking profits until the end of the day, so I'm not really sure if I could have done anything better.

    I've spent countless hours agonizing over trying to figure out ideal time to take profits based on my system. I'd like to think I have some sort of system, although I suspect that I need to do a lot of work in this area.

    Or is it not possible trying to improve on the take profit system? Also need to be careful to avoid jumping to conclusions based on a limited amount of recent trading datapoints.

    MTD core trading strategy up 24k
     
    #26     Jul 20, 2011
  7. do your strategies use technical analysis.. what is the typical trade duration of your core strategy?
     
    #27     Jul 20, 2011
  8. I would say its 99% fundamental analysis. Trade duration is intraday, enter at the start of the day (or aftermarket the prior day), and exit at the end.
     
    #28     Jul 20, 2011
  9. Qbit

    Qbit

    How long have you been doing this? got a job at the moment but I'm thinking of trying what you do sometime in the future.

    The thing is even getting single a Bloomberg terminal is pretty expensive and even if expecting e.g. a 10% annualized you need a pretty good amount already to 1) get a decent income out of this 2) reinvest something in your strategie(s).

    Did you have a lot of cash before you started this or family/friends support?

    Regarding programming, why don't you gradually add some VBA in your Excel backtests? after you learn VBA, writting up some VB.net should be easy to code up your automated system, unless of course you can only settle only for Citadel quality infrastructure :D
     
    #29     Jul 20, 2011
  10. What are you seeing for volatility?
     
    #30     Jul 20, 2011