New Members: Be sure to confirm your email address by clicking on the link that was sent to your email inbox. You will not be able to post messages until you click that link.

Some "new" indicators

I hear from a Larry Williams recent video on StockCharts that some additional Williams indicators are coming to StockCharts "soon". I would imagine before the end of the year. He's got some interesting indicators that are already on the site like the %R and Ultimate Oscillator. He's got some others that he discussed that could be great additions. I have no idea what, if anything, is actually coming.

Another "new" addition to the StockCharts Blogs is Shaun Downey and his blog called 4th Dimension. He's got some additional indicators that he mentions are coming to StockCharts. His indicators also look beneficial.

https://stockcharts.com/articles/downey/

Comments

  • lmkwinlmkwin ✭✭
    edited May 2020
    Arthur Hill has his own site now called TrendInvestorpro.com If you are not a member, you can access his older articles in the "Archives" for free. He still does a StockCharts show as well.

    I was watching an older video where he talked about using a Weighted Average indicator to assist in determining strength using multiple time frames.

    He "backtested" this concept on an all ETF portfolio and found the results to be good, buying the ones that were above a bullish threshold for the indicator and not owning the ones that were below the bullish threshold for the indicator. It worked well on long trades with a good reward/risk. He used the Stochastics for his weighted indicator.

    I set up this concept in the scan engine for the %R

    [Williams %R(25) *0.10 + Williams %R(50) *0.15 + Williams %R(75) *0.20 + Williams %R(100) *0.25 + Williams %R(125) *0.30]

    I haven't looked at the results intensively, but in concept, it does make sense. You can use it in the scan body by adding a > X or < X inside the last bracket so it's looking for a level. Or you can use it in the Rank By statement.

    The weightings are the *0.10 , *0.15, etc. The *0.10 is 10% so it only uses 10% of the shortest term value in assembling the weighted average. *0.30 means that it applies 30% of the longest term value into the average.

    You can play around with the indicator, the time frames, and the weightings. I just put the time and weightings that Arthur used in this statement. Those were the ones he backtested AND shared the results.
  • I ran this scan, which gets no syntax errors, and got results:

    [group is sp500]

    rank by [Williams %R(25) *0.10 + Williams %R(50) *0.15 + Williams %R(75) *0.20 + Williams %R(100) *0.25 + Williams %R(125) *0.30]

    But, the result comes out in ALPHA order, which means the rank by statement was not executed.
  • Checked your code and I get 500 results


  • very interesting.

    I ran it again - this time get a syntax error from the advanced scan editor (didn't last time), but no error from the "check syntax" button, and then results like yours - meaning a true ranking and not alpha by symbol.

    So, it looks like you are correct - some math is allowed. I don't know how to explain the differences in results. May have to do with interrupted internet transmissions? No idea.
  • I pulled the logic and code directly from an Arthur Hill video where he showed the scan code.

    I think I've mentioned before about getting intermittent errors/ oddities more and more since ACP was launched. I think it might be an update timing issue that is "resolved" over time, but tricky to discern at the time.

    I still think that your PPO suggestion was the better alternative for the question asked as it can be easily "verified" on a chart.
Sign In or Register to comment.