BioComp Profit 8 - Trading the ES

Discussion in 'Journals' started by chanelops, Nov 2, 2008.

  1. I've not been following this thread but I do know biocomp's products quite well.

    Profit 8 (or Dakota too) simply generate a signal. In the hands of skilled users the signals can be high quality enough to be robust over backtest and significant oos periods. That said, as a raw signal, intra-trade drawdown can be signficant. Why's that?

    1. Well, the simple fact that they trade eod is one inescapable factor.
    2. The average hold time for the master signal (final output) will have a major bearing on the scope for mid-trade drawdown.
    3. Biocomp's products don't have a strong risk management framework (they're not full featured trade management and backtest platforms). So, I recommend to anyone that they use other means to control risk with biocomp signals.


    Thx
    D
     
    #71     Apr 1, 2009
  2. In principal I agree but disagree in practice. If a system is robust but gets experiences some equity volatility due to the record volatility of the last few months what does that say about the system? In fact, is is possible for an eod system to not experience high equity volatility in today's environment without curve fitting?

    Therefore, what you're suggesting is likely to result in over fitting - unless you're talking about risk control which I've addressed in a separate response. But even then, risk control needs to be light touch and sensibly balance limiting of loss against maximization of gain.

    I do suspect that some of the people commenting on these signals don't have much risk capital and so are hyper-sensitive to drawdown...


    Thx
    D
     
    #72     Apr 1, 2009
  3. Really? So what does it use now? It used Meshes (nets by any other name), GRNNs, etc, etc. Or maybe the NN description just isn't considered sexy in trading these days :)

    {threadjack}
    As I have said before, my biggest beef with Biocomp SW is that they look and feel like they were developed by someone in their spare time in their garage - this is why when I did some NN work I ended using Matlab.
    {/threadjack}
     
    #73     Apr 1, 2009
  4. You're right Neural Nets are in there. Your condescending tone doesn't do you any favours though.
     
    #74     Apr 1, 2009
  5. Welcome to ET, I prefer to call it "Realistic Tone".
     
    #75     Apr 2, 2009
  6. Yes, I'm familiar with the downside of ET culture. Thankyou for needlessly introducing it to an otherwise unpolluted thread.
     
    #76     Apr 2, 2009
  7. With all due respect, but it would appear that what you imply as polution coincides with your mindless drivel. A statement was made about the apparent lack of NNs in the current version of Biocomp, I questioned that statement without actually attacking anyone (at the time) or passing judgment - unlike your statements and tone - asshole.

    Now how about you get of your high horse and get the discussion back on NNs and Biocomp and sheah - there's a good lad.
     
    #77     Apr 2, 2009
  8. :::Sighs:::

    You seem to be compulsively combative and rude for no credible reason. But let's leave it at that. I've had a look at some of your other posts and see that you're not the most mature contributor to these boards. My mistake for giving you the benefit of the doubt.

    Now, feel free to get in the last word if you must. I won't reply to you again.
     
    #78     Apr 2, 2009
  9. Nice try asshole. Perhaps you should cast your mind back to the EBTA thread and your "contribution" there "your highness" :p
     
    #79     Apr 2, 2009
  10. Chanelops,
    I've sent you a PM and haven't yet seen a reply. Have you missed it, or been away from the forum?


    Thx
    D
     
    #80     Apr 3, 2009