java.lang.Object
edu.cmu.tetrad.search.work_in_progress.Mmmb
All Implemented Interfaces:
IMbSearch

public final class Mmmb extends Object implements IMbSearch
Implements the Min-Max Markov Blanks (MMMB) algorithm as defined in Tsamardinos, Aliferis, and Statnikov, Time and Sample Efficient Discovery of Markov Blankets and Direct Causal Relations (KDD 2003).
Author:
josephramsey
  • Constructor Details

    • Mmmb

      public Mmmb(IndependenceTest test, int depth, boolean symmetric)
      Constructs.
      Parameters:
      test - The independence test used in the search.
      depth - The maximum number of variables conditioned on.
      symmetric - True if the symmetric algorithm is to be used.
  • Method Details

    • findMb

      public Set<Node> findMb(Node target)
      Searches for the Markov blanket of the node by the given name.
      Specified by:
      findMb in interface IMbSearch
      Parameters:
      target - The name of the target node.
      Returns:
      The Markov blanket of the target.
    • getPc

      public List<Node> getPc(Node t)
      Returns:
      a supserset of PC, or, if the symmetric algorithm is used, PC.
    • getAlgorithmName

      public String getAlgorithmName()
      Description copied from interface: IMbSearch
      The name of the algorithm.
      Specified by:
      getAlgorithmName in interface IMbSearch
    • getNumIndependenceTests

      public int getNumIndependenceTests()
      Description copied from interface: IMbSearch
      Number of independent tests.
      Specified by:
      getNumIndependenceTests in interface IMbSearch