How to make money if creating a new technical indicator ?

Discussion in 'Technical Analysis' started by Streetwise, Jul 30, 2017.

  1. %%
    Well ,we would hope the crowd would respect some of his work;
    he [Welles Wilder] likes his ADX better than RSI. But he noted @ the time he first published RSI, ,TA was real young, so most all were glad to get something, anything LOL..... I like his parabolic stop + reverse, [aka parabolic time price] but i dont use it much....................................................................................................
     
    #41     Aug 4, 2017
  2. I respect his work but it is still not an objective measure of data. End of. Personalising a pure factual debate is the height of sneak.
     
    #42     Aug 4, 2017
    murray t turtle likes this.
  3. %%
    RSI may have worked well the first year it was published ?? Dont really know .LOL:cool:
     
    #43     Aug 4, 2017
  4. Sprout

    Sprout

    Yes, I learned boolean algebra in High School. There's no more complex math required. Volume is the leading indicator of price. All price action math is based in the past.
     
    #44     Aug 5, 2017
  5. dartmus

    dartmus

    Price and volume occurring at the same point in time both share the same timestamp so in raw form they are equal. By saying volume leads price you're implying a relationship where volume predicts the future direction of price. The validity of this can be measured by using a correlation function and offsetting the index between the independent and dependent values.

    Assuming the existence of consistently applied rules then it's unnecessary for correlation to exist on every bar. It can still be used as a component of a viable strategy. I'm speaking about dry up volume which only occurs occasionally or any other volume condition or requirement that occurs infrequently or is required to occur in conjunction with other events such as price events. Conjunction can be as tight or loose as necessary. Meaning the events don't have to be simultaneous. Variables representing a range of time can be included and thus replace any fixed offset with a fuzzy offset.

    Anyone interested in the topic of "how to measure" if anything leads or lags price might find it worthwhile searching for leading correlation. I am emphasizing the words "how to measure" because previous discussions ignored the simple logic and bliss of offset correlation verses the (mind numbing) standard which is centered correlation (which is where the index of the indep and dep vars are identical). That discussion was interrupted or evolved into an argument about what could and could not be a leading indicator when I emphasized *anything* and everything has the potential to be a leading indicator.

    For instance, it was said (not by me) a moving average logically follows rather than leads price and it can't be any other way. I am in favor of good logic in the right place and at the right time. Like when formulating a simple adjustment to the standard correlation formula. Fortunately I've been blessed with enough (right brain) creativity to keep my logical (left brain) focused on what's important.
     
    Last edited: Aug 5, 2017
    #45     Aug 5, 2017
    d08, Sprout and Xela like this.
  6. Sprout

    Sprout

    Comments within quoted text.
     
    Last edited: Aug 6, 2017
    #46     Aug 6, 2017
    Xela likes this.
  7. dartmus

    dartmus

    Generally speaking,

    When adjustments to a strategy cause an increase in the Equity Curve, those adjustments point the way toward a leading indicator... but, ...monitoring an EC is an extremely crude method of mining for leading indicators. A correlation function enables narrowing that search.

    Most of the standard rules apply when using correlation. It's unwise to compare apples with oranges without first normalizing their values to a common scale such as 0 or 1 as determined by the conditional states of extremely dissimilar components.

    here's a link that describes the basics.
    http://whatis.techtarget.com/definition/correlation

    I'm unable to say more due to a complex web of numerous non-disclosure agreements.
     
    #47     Aug 6, 2017
  8. Sprout

    Sprout

    Ok, I understand more of which you speak. The concept that volume is the leading indicator of price comes from price informing when to measure volume. The measures of volume per timescale bar is distinct from the measures of volume as it goes through the order of events as a trend progresses through to maturation or interruption. A trend is defined by a dominant move of volume, a non-Dominant move of volume and a return to the original dominance. A trend can last a single bar before a new trend begins.

    When you describe normalized values to 1 and 0. Are you referring to reducing states down to Boolean values?

    Too bad about the non-disclosure agreements. They bind more than they free. I do respect your integrity in honoring the agreement, however there's not a fair-exchange of ideas if one party claims non-disclosure.

    It's based in a scarcity based paradigm that imprisons the mind.
     
    #48     Aug 6, 2017
  9. Bolean Bolan pahh.. red herring. Any indicator should be produced by objective scientific standards including volume, simple as that
     
    #49     Aug 8, 2017
    d08 likes this.
  10. Sprout

    Sprout

    Yes, true.

    Work that's ripe for you to do if you truly want to understand the market's system of operation.

    I've done the work and enjoy the fruits thereof. Countless others here on ET as well. They don't post a lot, too focused on purposeful learning by mining the archives of deep knowledge. These discussions by traders witnessed far greater changes in the marketplace and technology than you or I. They also have deeper wisdom and insight.

    Think whatever you want and enjoy those fruits. Post whatever makes you feel good.

    Understanding has no limits.
     
    #50     Aug 9, 2017
    Xela likes this.