Results 1 to 4 of 4

Math Help - Question about F-Test for comparing multiple regression models

  1. #1
    Newbie
    Joined
    Feb 2011
    Posts
    21

    Question about F-Test for comparing multiple regression models

    So the equation for comparing two multiple regression models using an F-test is



    When we're calculating this to compare our models, do we use the Sum of Sqaures for the model, or the Sum or Squares for the Error?
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Senior Member Sambit's Avatar
    Joined
    Oct 2010
    Posts
    355
    SS1 is the sum of squares of the factor whose effect you want to compare; and SS2 is the sum of squares of errors. Both SS1 and SS2 are calculated from the model. For example, if your model is y_{ij}=\mu+\alpha_i+e_{ij} and you want to test H_o: all \alpha_i's are equal; then your SS1 is \sum_{i,j} (\bar{y_{io}}-\bar{y_{oo}})^2 and SS2 is \sum_{i,j} (y_{ij}-\bar{y_{io}})^2
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Senior Member
    Joined
    Oct 2009
    Posts
    340
    You can actually do this by taking the difference in either the model sum of squares or the error sum of squares - the order flips if you use model SS - HOWEVER the error sum of squares (of the larger model) needs to be in the denominator.

    To be consistent with the way you've drawn up the test statistic, the SS terms are error sums of squares of their respective models.
    Follow Math Help Forum on Facebook and Google+

  4. #4
    Senior Member
    Joined
    Oct 2009
    Posts
    340
    Quote Originally Posted by Sambit View Post
    SS1 is the sum of squares of the factor whose effect you want to compare; and SS2 is the sum of squares of errors. Both SS1 and SS2 are calculated from the model. For example, if your model is y_{ij}=\mu+\alpha_i+e_{ij} and you want to test H_o: all \alpha_i's are equal; then your SS1 is \sum_{i,j} (\bar{y_{io}}-\bar{y_{oo}})^2 and SS2 is \sum_{i,j} (y_{ij}-\bar{y_{io}})^2
    This doesn't give a test of anything. Taking a difference in those two sums of squares doesn't give a sum of squares - it doesn't even give something that is positive w.p. 1. The question is about testing nested models in regression, not testing for factor effects in anova.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Ncaaf multiple regression/ stepwise regression
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: August 10th 2011, 06:21 AM
  2. Comparing model fits of non nested models
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: September 19th 2010, 06:19 AM
  3. Replies: 0
    Last Post: July 30th 2010, 12:21 AM
  4. Multiple linear regression: partial F-test
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: June 22nd 2009, 11:54 PM
  5. hypothesis testing comparing 2 multiple regression models
    Posted in the Advanced Statistics Forum
    Replies: 6
    Last Post: October 4th 2006, 01:52 PM

Search Tags


/mathhelpforum @mathhelpforum