docs.javahelp.manual.boxes.search.sem_bic.html Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of tetrad-lib Show documentation

There is a newer version: 7.6.6




    Scores: SEM BIC
    



    
    
        
            Score: SEM BIC
        
    
    

 Calculates a linear, Gaussian BIC score score for continuous variables.
    The score decomposes for a DAG into the sum of the BIC scores for each
    variable given its parents. The child is regressed onto its parents and
    a residual calculated. The likelihood is calculated as function of the
    standard deviation of this residual. The usual BIC penalty is multipled
    by a constant c which we call the 'penalty discount'. This parameter
    controls for the sparsity of the output.