Has anybody ever had a consistently profitable algorithm automated?

Discussion in 'Automated Trading' started by pmg, Jan 15, 2012.

  1. You must have a really low commission rate.
     
    #31     Jan 24, 2012
  2. One can't compete with on that timeframe/in that scale without. Lots of firms out there will offer hyper-low rates (relative to rates of the past, at least). You usually have to give up a percentage of profit (5-20%, but not always) and commit to some volume (2-10 million).

    A firm will give a very low rate out so long as they're making money; if a trader is paying $2-6k/month in fees, they'll happily give out a rate that may seem absurdly low (by the standards of the past) if, without such a rate, the trader wouldn't be able to work such a strategy and make any money (and therefore pay any commissions). Furthermore, the firm gets to get its cost of clearing lowered by having the volume. A lot of clearing deals start with a fixed minimum volume number which is not necessarily easy to meet with only 20 or 30 traders, so any trader paying even slightly more than the firm's cost is lowering the firm's overall cost and helping the firm make money.

    I don't want to hijack this thread and turn it into a commission/rate thread, nor will I respond to PMs about where I traded, what I paid, etc. s please don't ask.
     
    #32     Jan 24, 2012
  3. No, I do not think it is possible...are there other people here, I don't know, but the market is not random.

     
    #33     Jan 24, 2012
  4. The only consistent thing about the stock market is: "Thou shall not covet thy neighbor's property [nor his margin trading/investing account]".
     
    #34     Jan 24, 2012
  5. I decide to show up in this thread :). Yes there are people who are successful.
     
    #35     Jan 28, 2012
  6. I managed to put automated systems together , they made about 22 % a year over the last 10 years on back tests.From 2008 to 2010 the return dropped to 15 %.It is hardly worth running automated systems for 10 to 15 % a year , it simply is not attractive if additional slippage and broker execution problems are incurred , they systems could produce zero with the bucket shops.

    Automated systems can have one or two years of drawdowns and zero profits , how can anyone go through that period emotionally?

    We now operate the automated system for signals and trade some of them manually with decent brokers.

    The markets can be ranging or trading with few profitable trades for auto systems, thats why you will have to learnt to trade like us.
     
    #36     Jan 28, 2012
  7. oiler...please contact Baron if you would like to become a sponsor, otherwise knock it off....

    ES


     
    #37     Jan 28, 2012
  8. ===========
    Another problem/opportunity, is profitable trends are not predictions, automated or other wise.

    For example,GLD, gold does trend well, trends lengthly some ;
    but also one can find plenty of times[ years]when it does not trend well/profitable.


    :cool: Not to mention, we had some good discussions on GM;
    but that system/trade /invest is expired.LOL. Yes GM changed its name more or less, so have @ it again.:D
     
    #38     Jan 28, 2012
  9. Yes
    I do the programming myself
     
    #39     Jan 28, 2012
  10. I don't know if this has come up, but Strategy Runner and Strategy Exchange are hypothetical performance trackers of completely automated algorithms that you can subscribe to.

    There's quite a few of them, and they'd all fall into the consistently profitable algorithm automation category.
     
    #40     Jan 28, 2012