Class BicDiffPerRecord

java.lang.Object
edu.cmu.tetrad.algcomparison.statistic.BicDiffPerRecord
All Implemented Interfaces:
Statistic, Serializable

public class BicDiffPerRecord extends Object implements Statistic
Difference between the true and estiamted BIC scores.
Author:
josephramsey
See Also:
  • Constructor Details

    • BicDiffPerRecord

      public BicDiffPerRecord()
  • Method Details

    • getAbbreviation

      public String getAbbreviation()
      Description copied from interface: Statistic
      The abbreviation for the statistic. This will be printed at the top of each column.
      Specified by:
      getAbbreviation in interface Statistic
      Returns:
      Thsi abbreviation.
    • getDescription

      public String getDescription()
      Description copied from interface: Statistic
      Returns a short one-line description of this statistic. This will be printed at the beginning of the report.
      Specified by:
      getDescription in interface Statistic
      Returns:
      This description.
    • getValue

      public double getValue(Graph trueGraph, Graph estGraph, DataModel dataModel)
      Returns the difference between the true and estimated BIC scores, divided by the sample size.
      Specified by:
      getValue in interface Statistic
      Parameters:
      trueGraph - The true graph (DAG, CPDAG, PAG_of_the_true_DAG).
      estGraph - The estimated graph (same type).
      dataModel - The data model.
      Returns:
      The difference between the true and estimated BIC scores, divided by the sample size.
    • getNormValue

      public double getNormValue(double value)
      Returns the normalized value of the statistic.
      Specified by:
      getNormValue in interface Statistic
      Parameters:
      value - The value of the statistic.
      Returns:
      The normalized value of the statistic.
    • setPrecomputeCovariances

      public void setPrecomputeCovariances(boolean precomputeCovariances)
      Returns true if the covariances are precomputed.
      Parameters:
      precomputeCovariances - True if the covariances are precomputed.