+ Reply to Thread
Results 1 to 4 of 4

Thread: Logistic or OLS Regression?

  1. #1
    Points: 405, Level: 8
    Level completed: 10%, Points required for next Level: 45

    Posts
    6
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Logistic or OLS Regression?




    I am new to statistics and learning a bit about regression and logistic "regression."

    The dependent variable for my data set is whether a sports team wins, ties or loses (thus, three outcomes). There are multiple observations per team (1 per match). My first thought was multinomial logistic regression, but I've been having some problems with this in R. However, I am under the impression that since these are categorical results (ie 1, 2, or 3; loss, tie or win) that I cannot use ordinary regression.

    What are the statistical disadvantages in using ordinary regression instead of logistic regression in this case? Is it absolutely incorrect to use ordinary regression for this type of data set?

    Thanks in advance for your help!

    James

  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: Logistic or OLS Regression?

    ordinary regression = linear?


    Big picture:
    What would be the average of loss, tie, win? How would you interpret that? Think about least squares and plotting the observed and expected value (loss, tie, win). How would you do that? Would it be better to interpret the probability of groups?


    Most likely you are looking at multinomial.
    Stop cowardice, ban guns!

  3. #3
    TS Contributor
    Points: 12,227, Level: 72
    Level completed: 45%, Points required for next Level: 223
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,470
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: Logistic or OLS Regression?

    hi,
    I would say that it is not right to use OLS in this case. The numbers 1, 2, 3 are a little more then names ( though ordered), You could have a quite a different result if you coded the three outcomes 1, 1.5 and 1000 rather then 1 2 3 and there is no reason to prefer one over the other. So, Ordinal logistic regression seems to be the best way to analyze this type of data.

    regards

  4. #4
    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: Logistic or OLS Regression?


    If your dependent variable only has 3 levels you would probably want to use either ordered logistic regression if it was ordinal data or multinomial logistic regression if it is nominal.
    "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