+ Reply to Thread
Results 1 to 3 of 3

Thread: huge differences in R2 between OLS and ordered logit regression model

  1. #1
    Points: 1,050, Level: 17
    Level completed: 50%, Points required for next Level: 50

    Posts
    20
    Thanks
    3
    Thanked 0 Times in 0 Posts

    huge differences in R2 between OLS and ordered logit regression model




    I was wondering why I get huge differences in R2 when running a ordered logit model and a OLS model. I use the same DV and IVs for both models. The DV has values between 0-16. IVs are 5- point likert scale and dummy-variables.
    For the ordered logit odel, STATA gives me a R2<1%, whereas the OLS model indicates an R2 of 26%.

    Tanks in advance.

  2. #2
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: huge differences in R2 between OLS and ordered logit regression model

    It is calculated differently. One model based on ols the other probably ML
    Stop cowardice, ban guns!

  3. #3
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    Re: huge differences in R2 between OLS and ordered logit regression model


    The pseudo R2 of logistic regression have nothing in common with the R2 in OLS. They aren't calculated the same way and don't mean the same thing (there is no real logistic regression R squared equivalent of explained variance which is what r squared means in OLS). Also there are dozens of pseudo R squared in logistic regression and they are calculated differently and mean different things (often far different).

    It is doubtful whether it is even of value to use the logistic regression R squared. It was created because researchers liked having, or practisioners did, something like the OLS R squared. But it probably causes more confusion than good IMHO because people assume it means explained variance and it does not. Nor can researchers agree on how to calculate it.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

+ 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