Skip to main content
Figure 4 | BMC Bioinformatics

Figure 4

From: Assessing statistical significance in causal graphs

Figure 4

Computational complexity of Ternary Dot Product algorithms. Counts of the numbers of D-values computed by the simple quartic algorithm and during the thresholding part of the 2 × 2- and 3 × 2-family algorithms. Solid lines indicate total counts while corresponding dotted Lines indicate the numbers of contingency tables (respectively families) that pass the ϵDmax threshold. The left panel shows a "dense" case n0 = 5n+ while the right panel shows a "sparse" case n0 = 50n+. For these examples we set n+ = n- = q+ = q- and chose ϵ = 10-16.

Back to article page