Steven Carchon schrieb: > > Hello, > > During my statistical reading, I encountered the following : "the > factor model is generally not identifiable". What does this mean > actually and when is a factor model identifiable then ? What are the > conditions ? > > Many thanks, > Steven
I think, that you're recurring to the "rotational indeterminacy". To have a solution identifiable, you can add additional restrictions, like: rotational position must be in PCA-orientation, or in Varimax-orientation or should be according to a hypothesis about parts of the loadingsstructure. Gottfried Helms . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
