Changeset 180 for bdm/estim/mixef.h

Show
Ignore:
Timestamp:
10/15/08 19:08:06 (16 years ago)
Author:
smidl
Message:

Modifications of BDM to reflect changes in basics

Files:
1 modified

Legend:

Unmodified
Added
Removed
  • bdm/estim/mixef.h

    r177 r180  
    3333The 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. 
    3434 
    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 project this class itself belongs to the exponential family. 
     35This 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. 
    3636 
    3737TODO: Extend BM to use rvc. 
     
    4848        eprod* est; 
    4949        ////!Indeces of component rvc in common rvc 
    50  
     50         
    5151        //! Auxiliary function for use in constructors 
    5252        void build_est() { 
     
    9191        double logpred ( const vec &dt ) const; 
    9292        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); 
    9396}; 
    9497