NEWS.md
In AlexisDerumigny/ElliptCopulas: Inference of Elliptical Distributions and Copulas

ElliptCopulas 0.1.4

New functions (written with Victor Ryan @VictorRyan12 ) :
EllDistrDerivEst: nonparametric estimation of the derivatives of the generator of an elliptical distribution.
EllDistrEst.adapt: adaptive nonparametric estimation of the generator of an elliptical distribution.
estim_tilde_AMSE: estimate the component of the asymptotic mean-square error (AMSE) of the nonparametric estimator of the elliptical density generator that only depends on the parameter a.
EllDistrEst now works in a vectorized way, where a and/or h are vectors of the same length as the grid on which the estimator is computed. Each value of the grid is then estimated with the corresponding tuning parameters (corresponding element of a and of h).
New option averaging = "random" for the function KTMatrixEst corresponding to the averaging of a random set of entries in the off-diagonal blocks.
The output of KTMatrixEst now has colnames and rownames set to the names if available in blockStructure.

Fixed a bug in KTMatrixEst (whose output did not have ones on the diagonal, contrary to the documentation).
Fixed a bug in EllDistrEst when the variance matrix is not the identity.

Fixed a bug in EllDistrEst when mu is not zero. (#1, thanks to Rutger van der Spek)
EllDistrEst gains two new arguments: mpfr and precBits, that allows to use the package Rmpfr for multiple floating point precision (needed for dimensions larger than 250). (#2, thanks to Rutger van der Spek)
New function KTMatrixEst for fast estimation of Kendall's tau matrix, potentially under structural assumptions. (#2, thanks to Rutger van der Spek)
New dependencies: Import: Rmpfr, pbapply. Suggest: mvtnorm.