+ Reply to Thread
Results 1 to 6 of 6

Thread: R Squared and dummy variables

  1. #1
    Points: 4, Level: 1
    Level completed: 7%, Points required for next Level: 46

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    R Squared and dummy variables




    Hey,
    If you changed your model from one variable to a number of dummy variables, would the R^2 increase considerably, supposing that the dummy variables helped to create a more accurate estimate of the dependent variable?

    Thanks

  2. #2
    Omega Contributor
    Points: 38,413, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,004
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: R Squared and dummy variables

    Your question is not clear. So you are conducting linear regression and have a continuous predictor that you may discretize? Now your question is how it will impact the R^2 value for the model. Is that your question?


    We can talk about it, but I would recommend to first do it and report your findings. This will likely answer some of the basic question about what you are actually working with.

  3. #3
    Points: 657, Level: 13
    Level completed: 14%, Points required for next Level: 43

    Posts
    57
    Thanks
    1
    Thanked 9 Times in 8 Posts

    Re: R Squared and dummy variables

    It will affect the model fit, however there are a lot of arguments against dichotomising continuous variables...


    http://biostat.mc.vanderbilt.edu/wik.../CatContinuous

  4. #4
    TS Contributor
    Points: 17,779, Level: 84
    Level completed: 86%, Points required for next Level: 71
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,542
    Thanks
    56
    Thanked 640 Times in 602 Posts

    Re: R Squared and dummy variables

    If you changed your model from one variable to a number of dummy variables, would the R^2 increase considerably, supposing that the dummy variables helped to create a more accurate estimate of the dependent variable?
    I'd guess that this could happen in case of a nonlinear relationship between the predictor and the dependent variable.

    With kind regards

    K.


    @evelyn13
    Seemingly the OP didn't mean dichotomization, but multiple dummies.

  5. #5
    Omega Contributor
    Points: 38,413, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,004
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: R Squared and dummy variables

    Follow-up, I am trying to reference whether a relationship even exists in the first place.

  6. #6
    Points: 657, Level: 13
    Level completed: 14%, Points required for next Level: 43

    Posts
    57
    Thanks
    1
    Thanked 9 Times in 8 Posts

    Re: R Squared and dummy variables


    Quote Originally Posted by Karabiner View Post
    I'd guess that this could happen in case of a nonlinear relationship between the predictor and the dependent variable.

    With kind regards

    K.


    @evelyn13
    Seemingly the OP didn't mean dichotomization, but multiple dummies.
    True sorry...my points still stand though (including the link I attached).

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats