Changeset 180 for bdm/estim/mixef.h
- Timestamp:
- 10/15/08 19:08:06 (16 years ago)
- Files:
-
- 1 modified
Legend:
- Unmodified
- Added
- Removed
-
bdm/estim/mixef.h
r177 r180 33 33 The characteristic feature of this model is that if the exact values of the latent variable were known, estimation of the parameters can be handled by a single model. For example, for the case of mixture models, posterior density for each component parameters would be a BayesianModel from Exponential Family. 34 34 35 This class uses EM-style type algorithms for estimation of its parameters. Under this simplification, the posterior density is a product of exponential family members, hence approximate estimation projectthis class itself belongs to the exponential family.35 This class uses EM-style type algorithms for estimation of its parameters. Under this simplification, the posterior density is a product of exponential family members, hence under EM-style approximate estimation this class itself belongs to the exponential family. 36 36 37 37 TODO: Extend BM to use rvc. … … 48 48 eprod* est; 49 49 ////!Indeces of component rvc in common rvc 50 50 51 51 //! Auxiliary function for use in constructors 52 52 void build_est() { … … 91 91 double logpred ( const vec &dt ) const; 92 92 const epdf& _epdf() const {return *est;} 93 emix* predictor(const RV &rv); 94 //! Flatten the density as if it was not estimated from the data 95 void flatten(double sumw=1.0); 93 96 }; 94 97