trend following delusion shattered

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

Thread Status:
Not open for further replies.
  1. Here's a free (but risky) donut from me:

    I can only place 2 cents this bet, as this model could be practically very wrong relatively to many more experienced and knowlegeble traders. I consider myself should be too old to do this kind of formal analysis, but anyway here is my trial.

    dV = Vb - Va

    dV (change of value)
    Va (starting value at time a with price Pa)
    Vb (ending value at time b with price Pb)

    Vb = Sum (Vbu, Vbd)
    Where u for price up, d for down

    Vbu = Expected return if Pb is up = Pbu1*Cu1 + Pbu2*Cu2 + ... + Pbun*Cun

    where Cu is the chance/ probability of the price Pb if going up, i.e. Pbu

    Cu1 + Cu2 + ... + Cun =1,

    Pbu1 is the price happening corresponding to Cu1,

    1, 2, ..., n are the number of possibilities happening at Pbu;


    Vbd = Expected return if Pb is down = Pbd101*Cd101 + Pbd102*Cd102 + ... +Pbdn*Cdm

    Cd101 + Cd102 + ... + Cdm = 1,
    where Cd is the chance/ probability of the price Pb if going down, i.e. Pbd

    Pbd101 is the price happening corresponding to Cd101,

    101, 102, ..., m are the number of possibilities happening at Pbd;.

    The above model is a very simplified and basic one, mainly for the sake of this thread for discussion purpose, thanks to Nickel and Hank, besides others. Hence, I have to warm readers that using it for any trading activities can be very risky.

    :confused:
     
    #631     Mar 23, 2005
  2. That above lines should be changed to:

    R = dV/ Va
    dV = Vb - Va

    R (return of holding from time a to time b)
    dV (change of value)
    Va (starting value at time a with price Pa)
    Vb (ending value at time b with price Pb)

    :confused:
     
    #632     Mar 23, 2005
  3. Perhaps a text version of the above model can be referred to "Chapter 7, The Trader's Edge: Thinking in Probabilities" , Trading in the Zone, by Mark Douglas.

    Q
    "It's the abilities to believe in the unpredictability of the game at the miscro level and simultaneously believe in the predictability of the game at the macro level that makes the casino and the professional gambler effective and successful in what they do.

    ...

    The bottom line is that there is some degree of sophistication to thinking in probabilities, which can take some people a considerable amount of effort to integrate into their mental systems as a functional thinking strategy. Most traders don't understand this; as a result, they mistakenly assume thay are thinking in probabilities, because they have some degree of understanding of the concepts."

    ...

    Traders who have learned to think in probabilities are confident of their overall success, because they commit themselves to taking every trade that conforms to their definition of an edge.

    ...

    They completely accept what the market is offering them, and they wait for the next edge."

    UQ

    :confused:
     
    #633     Mar 23, 2005
  4. Just a simple (and perhaps a stupid) question:
    Can we pass on to the reality instead of discussing theoretically?

    What’s the use of finding out if the d of the p-f for a series of observations that have no negative correlation with the square of the standard deviation of the mean of the 90% of the most reliable data of the observation is representative as basis for a prediction of the maximum deviation of the probable future evolution of the quotes, without forgetting the stand of the moon and the exact date on the Chinese calendar?

    Or in simple human language: just post track records, because after all, that’s what it is all about, not?

    :D :D :D :D
     
    #634     Mar 23, 2005
  5. It's your turn now. Your win-rate is over 91%, whereas mine below 19%. :confused: :D

    PS: btw, would you like to present your model as well? :confused:
     
    #635     Mar 23, 2005
  6. I can publish several models, but the real good stuff i will never publish, i'm not that idiot.

    I will perhaps once publish all the trades from 1 week to give you an idea of my returns. But than will happen what happened before, there will be remarks in an attempt to break it down; remarks like:
    -you probable had several accounts and took the only profitable one to show us
    -you had two accounts and took each time the opposite trade, afterwards you showed us the best one
    -can the track record be certified and are you, and not someone else , the one who traded
    -you had luck, can you prove your abilities with a track record of 25 years (but i can have 25 years of luck also)

    And all the proof will be put in discussion again, with arguments like those as we see here for more than 100 pages now.

    So the best thing to do is trade, don't publish anything, and read this thread every now and than to have a good laugh about trends that seem to be non existent.


    PS: how can you make a profit with only 19% win rate? You must be taking the huge trends then.
     
    #636     Mar 23, 2005
  7. Do trends exit? :confused:
     
    #637     Mar 23, 2005
  8. I don't break my head on that one. I know that profit exists, that's enough to me. If it comes from trend, non trend, good luck, bad luck or no luck at all doesn't matter to me anymore. I like to keep things simple.

    You might find the answer to your question in the previous 100+ pages. Or you might be more confused afterwards.


    :D
     
    #638     Mar 23, 2005
  9. Good thinking. :)
     
    #639     Mar 23, 2005
  10. Whomever taught you in high school & college, needs spanked.
     
    #640     Mar 23, 2005
Thread Status:
Not open for further replies.