Chop vs. Trend

Discussion in 'Automated Trading' started by Yag, May 1, 2005.

  1. ssrrkk


    there is this thing called a STFT -- short-term fourier transform used sometimes in dynamic speech processing...
    #41     Jun 12, 2005
  2. ssrrkk


    also, the stationarity constraint is precisely the caveat I mentioned in my prev post (the need for your estimated spectrum from the immediate past to be close to that for the duration of time in the future during which you will trade)...
    #42     Jun 12, 2005
  3. :D :D :D
    #43     Jun 12, 2005

  4. I agree, weekly Dow ADX has been under 20 for 18 months. not much trend their. The longer it stays under 20 the better the next trend will be. The daily Dow ADX has been falling from over 40 for the last 6 weeks, its now under 15 as the Dow has been in a tight range for the past 3 weeks. These 70-100 point intraday swings are accumulation or distribution. something will happen soon IMO.

    So for now just try and scalp some profits intra-day. and be happy.
    #44     Jun 12, 2005
  5. I tend to agree insofar as exploitable trend identification is concerned. Sometimes, however, I get a transient feeling that I'm just being lazy and that I should hunker down and try to figure out the math of trend identification in a way that it can be reliably exploited rather than only identified after the fact. When that happens, I usually just take a few deep, healthful, cleansing, life-affirming breaths of fresh air and the feeling miraculously goes away. It's really quite refreshing.
    #45     Jun 12, 2005
  6. LMeyers


    Yes, I think so too. I think that the indices are getting ready to start trending bigtime - to the upside.
    #46     Jun 12, 2005
  7. (i) What makes you believe that there is a "math of trend identification"?

    (ii) If such a "thing" would exist, what makes you believe that you could "figure it out"?

    #47     Jun 12, 2005
    Oysteryx likes this.
  8. ssrrkk


    how about this. use your favorite spectral analysis tool to find the dominant wavelengths of your timeseries. ifyou do find long wavelength components that have much larger coefficients compared to your high-frequency components (i.e., you must screen for stocks with this property), then you might be able to choose the appropriate window for your moving average indicator... pretty basic idea that should be easy to back test.
    #48     Jun 12, 2005
    Oysteryx likes this.
  9. slacker


    At the end of the day you will have results that are only backfitted to the data series.

    There is nothing in Amp, phase or frequency that will help you trade the next bar. Any research to the contrary would be great news however... Have you had success applying signal processing techniques to market data so that you can trade the signals?

    Thank you.
    #49     Jun 12, 2005
    Oysteryx likes this.
  10. I hope your right, I only trade from the long side. Put Call sure is low though for a big rally. seasonality is not that favorable either.

    Alll that being said im sure we just could rally.
    #50     Jun 12, 2005