Mathematical expectation

Discussion in 'Risk Management' started by Visaria, Jul 30, 2014.

  1. Visaria

    Visaria

    Not scared, mate! Will pick it back up when i get back. Thanks again, Mav :)
     
    #21     Jul 30, 2014
  2. panzerman

    panzerman

    If you have a look at John Ehlers latest book "Cycle Analytics for Traders", in chapter 17 there are two sections which may help you called "Evaluating a Trading Strategy" and "Monte Carlo Evaluation." In fact you could Google it and probably find those sections online.

    Also, I got this spreadsheet from a John Ehlers article in Stocks and Commodities magazine, haven't used it yet though.
     
    #22     Jul 30, 2014
  3. kut2k2

    kut2k2

    You could have said that up front.
    Only if you believe in fixed trade distributions. If so, I must ask if you also believe in Santa Claus and the tooth fairy. The phrase 'trade distribution' is very misleading. It gives the impression of some tangible thing despite a complete absence of evidence of its existence. To put it another way: properly calculated, your optimal trading fraction is a moving target.
     
    #23     Jul 30, 2014
  4. Sergio77

    Sergio77

    Relying on expectancy is like sitting in an empty bar waiting for a supermodel to come in and talk to you. Expectancy says what happened in the past. The future may and will be different. Some authors who have traded little or maybe not at all have promoted this statistic and many clueless traders think it is something useful. The only useful statistic is whether one makes money. Expectancy may be high and still disaster can strike. and it will.
     
    #24     Jul 31, 2014
  5. Maverick74

    Maverick74

    Actually that is not what expectancy says...
     
    #25     Jul 31, 2014
  6. SIUYA

    SIUYA

    I guess for most of us we can expect very little as this only happens in the imagination....not in the past or the future.
     
    #26     Aug 1, 2014
  7. ronblack

    ronblack

    +1

    Another great post from sergio77. It is actually worse than that as I believe that expectancy is another silly concept invented by author non-traders. One outlier event and the expectancy changes by a large percentage. It is totally useless to average past trades in an effort to see if there is an edge. People learned that the hard way in the 1987 crash, in the 200 dot-com bust and in the 2008 financial crisis collapse. The most useful measure is "robustness" but it is not well defined. Taleb's "antifragility" is a better one against those silly expectancy claims but hard to quantify. Anyone who relies on expectancy will lose, I tell you that much.
     
    #27     Aug 1, 2014
  8. Maverick74

    Maverick74

    Ron, expectancy is not used to determine if one has an edge.
     
    #28     Aug 1, 2014
  9. ronblack

    ronblack

    How so Maverick? Negative expectancy means no edge. Therefore, and directly so, expectancy has been used to determine, maybe is more correct to say, the possibility of an edge. The effort to do that goes beyond expectancy but it starts from there. The point I'm trying to make is that expectancy is the wrong starting ground, it is useless. Many systems with negative expectancy 5 years ago have now turned profitable and many with positive are losing constantly. Expectancy does not stay constant but fluctuates around zero line. Some authors that made a carrier out of this concept of expectancy will never tell you this because maybe they do not even understand it. It has to, otherwise a little (or big) guy with positive constant expectancy could own the world in a few years. Two important things:

    - Expectancy always fluctuates about 0
    - Expectancy is not scalable

    The two combined make this a useless concept. Anti-fragility and robustness are more useful concepts. The sooner one understands this, the better for his money.
     
    #29     Aug 1, 2014
  10. imo,

    Expectancy = (Probability of Win * Average Win) – (Probability of Loss * Average Loss)

    Edge exists, If {[Expectancy/(Number of Trades)] - (Cost of Options per Trade) where [(Cost of Options)<(Premium of Options)]} >0

    LOL

    PS: Edge can exist in other forms!
     
    #30     Aug 1, 2014