NEWS

---

matrixCorr 0.12.3 (2026-09-28)

New features

• Distance-correlation API correction: dcor() now returns the conventional sample distance correlation (R_n). Previous versions returned a U-centred bias-corrected squared distance-correlation statistic. The latter is now exposed as bcdcor(), with signed finite-sample values preserved. Use dcor(..., squared = TRUE) for conventional (R_n^2).
• Added robust_ccc() for pairwise MCD-based robust concordance correlation. The estimator substitutes joint reweighted minimum covariance determinant location and scatter estimates into Lin's concordance coefficient and exposes alpha, mcd_nsamp, and seed controls.
• robust_ccc() supports the package's standard matrix, sparse, and edge-list outputs; error, complete, and pairwise missing-data modes; and the shared print(), summary(), plot(), estimate(), coef(), tidy(), ci(), and confint() interfaces.
• Optional paired percentile-bootstrap confidence intervals are available via ci = TRUE, with bootstrap success counts retained in result diagnostics.