Excel Divergence Formula

Discussion in 'Trading Software' started by gemini_315, Dec 15, 2004.

  1. gemini_315

    gemini_315 Guest

    I am looking for a formula to detect divergences between an indicator and price eg. RSI and price. I need to do this in Excel. If anyone has any info, I would be very grateful.
     
  2. gemini_315

    gemini_315 Guest

    Just bumping this....

    I am looking for a formula to detect divergences between an indicator and price eg. RSI and price. I need to do this in Excel. If anyone has any info, I would be very grateful.
     
  3. DVB

    DVB

    Try building a regression on both RSI and the price. Use the SLOPE function in Excel. If one has a positive slope, and the other one has negative one, you have a result. The only problem is that you have to define your time frame, but you can do a couple of formulas for different periods.

    Good luck

    DVB
     
  4. ToTrade

    ToTrade

    Does anybody have a formula for this?
     
  5. Listen-up guys: this is pretty complex. I've seen some Easy Language code HUNDREDS OF LINES long attempt to do this.
    Bottomline: It ain't easy.

    See this guy, who has spent a lot of time researching this subject:

    Inside Edge Systems is managed by Bill Brower. You can contact him several ways as follows:

    Tel: 203-454-2754
    Fax: 203-286-1321
    email: 1000mileman@mindspring.com
     
  6. Here's an idea about how you can assemble one. The conv/div indicator simply looks at the %change slopes over a fixed period. When they are in sync (convergent) it is green and positive, when they are out of sync (divergent) it is red.

    Keep in mind, like all other TA, it is subjectively interpreted. But the math is objective. Here is a 5 period example.
    Notice some divergent periods that were outlined.

    <img src="http://elitetrader.com/vb/attachment.php?s=&postid=1524701" border="0" alt=""><br /></font></p>
     
  7. By the way, this graph indicates a perfect example of how ta itself is very subjective.
    Notice how the RSI and Nasdaq were both mathematically convergent from 10/06 right up to 11/06, the market pulled back, but there was no divergent signal. No matter how you fine tune the period of the indicator, there will be times when it outright fails to predict future moves. I haven't done longer periods, but they would likely be slightly more predictive and robust.

    ---------------------------------------------------
    "Patterns are the fool's gold of the financial markets... They are the inevitable need for humans to seek to find patterns in the patternless." Mandelbrot