Uses of Class
edu.cmu.tetrad.search.utils.EdgePriors
Packages that use EdgePriors
Package
Description
This package contains classes for scoring causal graph models.
This package contains classes for testing causal graph search algorithms.
This package contains utility classes for causal graph search algorithms.
-
Uses of EdgePriors in edu.cmu.tetrad.search.score
Constructors in edu.cmu.tetrad.search.score with parameters of type EdgePriorsModifierConstructorDescriptionEdgePriorScore(Score score, EdgePriors priors) Constructs a prior-adjusted score. -
Uses of EdgePriors in edu.cmu.tetrad.search.test
Constructors in edu.cmu.tetrad.search.test with parameters of type EdgePriorsModifierConstructorDescriptionEdgePriorTest(IndependenceTest test, EdgePriors priors, double alpha) Constructs a prior-adjusted test. -
Uses of EdgePriors in edu.cmu.tetrad.search.utils
Methods in edu.cmu.tetrad.search.utils that return EdgePriorsModifier and TypeMethodDescriptionstatic EdgePriorsEdgePriors.fromMatrix(List<String> names, double[][] values, EdgePriors.Semantics semantics) Builds a store from a dense symmetric matrix and the variable names it is indexed by.static EdgePriorsEdgePriors.fromMatrix(List<String> names, double[][] values, EdgePriors.Semantics semantics, boolean allowZeroWeights) Builds a store from a dense symmetric matrix, optionally permitting zero weights.static EdgePriorsEdgePriors.neutral(EdgePriors.Semantics semantics) Returns an empty store: every pair takes the neutral value.EdgePriors.normalizedToMeanOne()Returns a copy of this WEIGHTS store rescaled so that the stored weights average one.EdgePriors.restrictTo(List<Node> variables) Returns a copy of this store keeping only the pairs both of whose variables occur among the given ones.EdgePriors.toWeightsAnchoredAtAlpha(double alpha) Converts prior log-odds to p-value weights anchored at a given significance level, so that beta_ij = 0 reproduces that level exactly.EdgePriors.toWeightsViaBicBridge(double lambda, int n, double alpha) Converts prior log-odds to p-value weights via the BIC bridge.