plotROCCurve: Plots ROC curve from estimated and true causal effects

In cnowzohour/greedyBAPs: Greedy BAP Learning Using Penalised Maximum Likelihood Score

Description

Plots ROC curve from estimated and true causal effects

Usage

plotROCCurve(CE.gt, CE.greedy, tol = 1e-08, precision.recall = FALSE,
  avg.auc = TRUE)

Arguments

CE.gt	List of minimal absolute causal effect matrices E (one per simulation run) for the ground truths, where (E)_i,j is the causal effect on j on i (as returned by causalEffectsSimulation)
CE.greedy	same for the greedy search results
tol	Tolerance for causal effects (everythin smaller than this will be thresholded)
precision.recall	Plot precision-recall instead of FPR / TPR?
avg.auc	If TRUE: Average of AUCs of individual curves If FALSE: AUC of averaged curves

Value

List of three elements: FPR - Matrix with false positive rates for each of the simulation runs (one column per run) TPR - Matrix with true positive rates for each of the simulation runs (one column per run) auc - Averaged AUC

cnowzohour/greedyBAPs documentation built on May 13, 2019, 1:36 a.m.