Hello,
I am new to fmm. If I run a two component mixture model I will have two sets of predicted values (one for each component). I can also generate the posterior prob and the most likely latent class membership. How do I combine these two sets of values to produce single predicted values which correspond to my dependent variable?
Thanks
Consider and compare (1) for any data point, use the predicted value for its most likely latent class. (2) value for class 1 * p(class 1) + value for class 2 * (1 - p(class 1)),
NB: p(class 2) = 1- p(class 1). I assume you added covariates to improve your prediction of class membership.
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwjro5_CzbHtAhWND2MBHaV8B8oQFjAAegQIBRAC&url=https%3A%2F%2Fwww.researchgate.net%2Fpost%2FInterpreting_output_of_predicted_values_in_a_finite_mixture_model&usg=AOvVaw30Xu8kzRhWbuMcKeZ2t6eT
- Badges
- Science topic