arXiv (Cornell University) · 2016 · 11 citations · 0 references
Natural SciencesMathematicsProbability TheoryStatisticsStochastic ProcessesInferential StatisticsStatistical InferenceIndependenceMultivariate Random Variables
Bergsma and Dassios (2014) introduced an independence measure which is zero if and only if two random variables are independent. This measure can be naively calculated in $O(n^4)$. Weihs et al. (2015) showed that it can be calculated in $O(n^2 \log n)$. In this note we will show that using the methods described in Heller et al. (2016), the measure can easily be calculated in only $O(n^2)$.