Class ArrowheadFn

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

public class ArrowheadFn extends Object implements Statistic
The arrow precision. This counts arrowheads maniacally, wherever they occur in the graphs. The true positives are the number of arrowheads in both the true and estimated graphs. Thus, if the true contains X*->Y and estimated graph either does not contain an edge from X to Y or else does not contain an arrowhead at X for an edge from X to Y, one false positive is counted. Similarly, for false negatives.
Author:
josephramsey
See Also:
  • Constructor Details

    • ArrowheadFn

      public ArrowheadFn()
      Constructs the statistic.
  • 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)
      Description copied from interface: Statistic
      Returns the value of this statistic, given the true graph and the estimated graph.
      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 value of the statistic.
    • getNormValue

      public double getNormValue(double value)
      Description copied from interface: Statistic
      Returns a mapping of the statistic to the interval [0, 1], with higher being better. This is used for a calculation of a utility for an algorithm.If the statistic is already between 0 and 1, you can just return the statistic.
      Specified by:
      getNormValue in interface Statistic
      Parameters:
      value - The value of the statistic.
      Returns:
      The weight of the statistic, 0 to 1, higher is better.