Feddback Regression

Optuma Forums Feature Requests Feddback Regression

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
    Posts
  • #57961
    AvatarThomas
    • Topics: 59
    • Replies: 84
    • Posts: 143

    Hi Mathew,

    first of all I want thank you really very much for your tireless efforts to make Optuma still better. I am very curious about v1.6 with all its new tools and improvements.

    Concerning your video about the use of regression, correlation, alpha, and beta I am very pleased to hear that you make the current somewhat confusing calculations about r-squared to a consistent and comparable one.

    In this regard I want mention for those interested in this topic a forum post in which we have already discussed this matter intensively. The link is: https://forum.optuma.com/topic/regression-r-squared/

    Concerning “beta” I want ask you Mathew whether you have also changed the calculation from normal percentage calculation to LN calculation as I have described it in the post “Beta #2”? The link to this post: https://forum.optuma.com/topic/beta-2/

    Thanks again for your great efforts to make Optuma the best technical and quantitative analysis program available.

    Thomas

    #57965
    AvatarThomas
    • Topics: 59
    • Replies: 84
    • Posts: 143

    Addendum: The video about regression, correlation, alpha, and beta was published on Twitter.

    The Link: https://twitter.com/Optuma

    Thomas

    • This reply was modified 3 weeks, 2 days ago by AvatarThomas.
    #57969
    MathewMathew
    • Topics: 28
    • Replies: 1,335
    • Posts: 1,363

    Hi Thomas,

    Thanks for the reminder – I still need to look into this.

    Since Beta is the slope of the regresion line, and all four values (correlation, alpha, beta, and r²) are components of the same regression calculation, they would all use the Ln instead of % change. I believe that is the expected behaviour but please let me know if you think otherwise.

    All the best

    Mathew

     

    #57971
    AvatarThomas
    • Topics: 59
    • Replies: 84
    • Posts: 143

    Hi Mathew,

    that is exactly my thinking. All calculations should use the Ln instead of the % change calculations.

    Thanks

    Thomas

    #57977
    MathewMathew
    • Topics: 28
    • Replies: 1,335
    • Posts: 1,363

    Ok – example of the new calculations on the Beta Post https://forum.optuma.com/topic/beta-2/

Viewing 5 posts - 1 through 5 (of 5 total)
  • You must be logged in to reply to this topic.

Pin It on Pinterest

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.

Close