trend following delusion shattered

Discussion in 'Trading' started by hank rollins, Mar 15, 2005.

Thread Status:
Not open for further replies.


  1. i am certainly not saying that you don't have an edge. obviously, if you can make a living at this game, you edge is working.

    did not van tharp show that one can make money by using random entries and good risk management?

    , i am saying, that trend following as talked about in the books about such things is not an edge--regardless of how the past makes it look.

    best,
    surfer
     
    #1281     Mar 13, 2006
  2. #1282     Mar 13, 2006
  3. kut2k2

    kut2k2

    Can't do a google search? Anyway ...

    http://www.trendfollowing.com/whitepaper/Does_trendfollowing_work_on_stocks.pdf
     
    #1283     Mar 18, 2006
  4. cnms2

    cnms2

    #1284     Mar 18, 2006
  5. Im not sure about longer term trading but on an intaday basis I often make substantial wagers that the trend will continue in situations like this:

    <img src="http://www.elitetrader.com/vb/attachment.php?s=&postid=1013113">

    Therefore I have not the slightest doubt that the odds are better than 50/50 the trend will continue after certain chart formations.(The link below is the same chart posted above)
     
    #1285     Mar 18, 2006
  6. fwiw, i do make a living trading futures.

    i do NOT use trend following techniques (on an intraday basis) GENERALLY speaking.

    however, this is not to say that trend following does not work - in certain markets.

    even neiderhoffer (and others) have conceded that trend following works in some markets.

    on an intraday futures scalping method, i am much more likely ot be fading moves (countertrend) vs. FOLLOWING a trend, but given proper market internals, signals, etc. sometimes i am "trend following" given the proper signals, market internals, etc.

    as an example of a trend following market, look @ the nikkei right now.

    the thing is making such hyooge intraday moves day after day that it is clearly tradeable on a trend following method. a simple SMA cross is enuf. Add some confluency with support/resistance levels, order flow monitoring, and some key price levels and it gets even better

    a counterexample would be the indexes yesterday (triple witching). those things were choppier than lizzie borden. but that's to be expected.

    markets are ALWAYS trending, but in MP parlance - they are usually trending horizontally. :) not vertically.
     
    #1286     Mar 18, 2006

  7. i am familiar with that sales pitch disquised as a scientific paper.

    unfortunately, upon careful reading, one can easily see the motivation of the authors to skew the results to support their tactics.

    sorry, the paper contains several flaws and is lacking in true unbiased research.

    best,

    surfer
     
    #1287     Mar 18, 2006
  8. cnms2

    cnms2

    Would highlight some of those flaws? It seemed ok to me ...
     
    #1288     Mar 18, 2006
  9. ecritt

    ecritt

    Yes please elaborate on the "flaws" and how the results were "skewed".
     
    #1289     Apr 16, 2006
  10. nbates

    nbates

    Have to agree, not much science there.

    Google is a great tool (tip - use "filetype;pdf" after criteria, and colon instead of semicolon...makes smiley here in the post so had to replace)

    Dynamical Modeling with Kernels for Nonlinear Time Series Prediction
    http://eprints.pascal-network.org/archive/00000484/01/NIPS2003_AA17.pdf

    Intraday Patterns and Local Predictability of High Frequency Financial Time Series
    http://molgedey.de/publications/MolgedeyEbeling_IntradayPaterns.pdf

    ps. these aren't anything to live-by, but good data and legitimate research none the less.
     
    #1290     Apr 16, 2006
Thread Status:
Not open for further replies.