Abstract
In an extension of Kendall’s (Formula presented.) , Bergsma and Dassios (Bernoulli 20(2):1006–1028, 2014) introduced a covariance measure (Formula presented.) for two ordinal random variables that vanishes if and only if the two variables are independent. For a sample of size n, a direct computation of (Formula presented.) , the empirical version of (Formula presented.) , requires (Formula presented.) operations. We derive an algorithm that computes the statistic using only (Formula presented.) operations.
| Original language | English |
|---|---|
| Pages (from-to) | 315-328 |
| Number of pages | 14 |
| Journal | Computational Statistics |
| Volume | 31 |
| Issue number | 1 |
| DOIs | |
| State | Published - 1 Mar 2016 |
| Externally published | Yes |
Keywords
- Binary tree
- Kendall’s tau
- Nonparametric correlation
- Rank correlation
- Spearman’s rho
- Test of independence
Fingerprint
Dive into the research topics of 'Efficient computation of the Bergsma–Dassios sign covariance'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver