|
|
Comparing Linear and Quadratic GMMs |
|
Message/Author |
|
|
Is it possible to use the standard chi-square difference test to compare linear and quadratic growth mixture models with the same number of classes? For example, can I use the chi-square difference test to say whether the two-class linear model is better than the two-class quadratic model? |
|
|
I found the topic "Decide linear or quadratic for GMM model" (I swear I searched the board before posting) and it seems to suggest that you should not use the chi-square difference test but should use BIC instead. Is that correct? |
|
|
Use BIC. |
|
Back to top |
|
|