|
|
 |
| GMM with diff. shapes (linear, quadra... |
 
|
| Message/Author |
|
|
|
|
Hi,
is it possible to estimate a GMM in which the trajectories of one class follow a linear shape, in a second class a quadratic shape and in a third class a s-shape?
Thanks.
Best regards,
Johannes |
|
|
|
|
| You should specify the most general growth model in the overall part of the model and make modifications in the class-specific parts of the model. For example, if you specify a quadratic model in the overall part of the model and want a linear model for one class, fix the mean and variance of the quadratic growth factor to zero in that class. |
|
|
| Back to top |
|
|
