NEWS


tlrmvnmvt 1.1.2 (2022-06-09)

This version revised the F77 callings to BLAS and LAPACK functions to satisfy the new requirement since R4.2.0.

tlrmvnmvt 1.1.1 (2022-01-28)

This version revised the argument check. Previously, setting the lower bound as a single value (not a vector) would cause error for the tile-low-rank functions. Now this problem is fixed.

tlrmvnmvt 1.1.0 (2020-04-06)

This version makes changes to the R interfaces of the package 1. The previous pmvn.genz and pmvn.tlr functions are combined into the new pmvn function 2. The previous pmvt.genz and pmvt.tlr functions are combined into the new pmvt function 3. The interfaces of the pmvn and the pmvt functions resembles those of the pmvnorm and the pmvt functions from the mvtnorm package at the best effort. Differences include: 1. Input correlation matrices are treated the same as input covariance matrices 2. Users can specify whether the results should be returned in terms of logarithm 3. The sample size, block size (for tile-low-rank only), and the truncation level can be specified through the algorithm parameter 4. Covariance matrices can be constructed by the package with the optional parameters but these parameters are not specified to align the function interfaces with those from the mvtnorm package

tlrmvnmvt 1.0.1 (2020-01-11)

This version fixed four bugs: 1. The memory allocation size error in functions mvn_internal, mvn_internal2, mvt_internal, and mvt_internal2 2. Change the position of the memory release command 3. Correct the functioning of the mean parameter. It was assumed to be "shifted" but should be "Kshirsagar"; Refer to mvtnorm::pmvt function's documentation for the meaning of the two names 4. Implement the internal Matern function so that it produces the same result as the geoR::matern function

tlrmvnmvt 1.0.0 (2019-11-04)

The initial release of the package.

For a complete description of the package functions, please refer to the help file. To install a highly optimized version of this package, please refer to the README.md file.