NEWS
TruncatedNormal 2.3.1
Bug fixes
rtmvt correctly samples when a loc vector is provided.
TruncatedNormal 2.3 (2024-07-08)
Bug fixes
rtnorm now checks arguments length and throw an error rather than try to recycle mean and standard deviation vectors (issue #6).
rtnorm recycles arguments when ltrunc or rtrunc are length 1.
*tmvt and *tmvnorm now check that the scale matrix sigma is symmetric, positive-definite and non-degenerate (issue #8) by default.
rmvnorm returns a vector rather than an 1 by n matrix for the unidimensional setting.
- Degrees of freedom larger than 350 are not treated as infinite, as the code suffers from overflow and returns missing values otherwise.
Changes
- The package uses
tinytest rather than testthat for unit tests.
- Since scrambling is disabled in
randtoolbox (throwing warning messages), the package now relies on the qrng package and the Owen scrambling method from spacefillr package for quasi Monte Carlo.
- Added alias in package documentation, as requested by Kurt Hornik.
TruncatedNormal 2.2.2 (2021-09-08)
Bug fixes
- Remove ghost argument
log (never implemented) from functions pmvnorm and pmvt; these now return a warning (thanks to Elysée Aristide)
TruncatedNormal 2.2 (2020-05-17)
Bug fixes
ptmvnorm and ptmvt now return 1 if all points exceed the upper bounds (incorrectly returned zero in v2.1.2 and NA in previous versions) (thanks to Adelchi Azzalini).
- Probability estimates now capped at 0/1 in ptmvt/ptmvnorm.
- Check for solution to saddlepoint problem; the code now considers convex constrained optimization program for normal samples (QMC or MC).
Changes
- No more warning for univariate Student distribution function evaluation.
- Unit testing for wrappers.
TruncatedNormal 2.1.2 (2020-05-02)
Bug fixes
ptmvtnorm and ptmvt return 0 for points outside the truncated region instead of NA (thanks to Adelchi Azzalini)
- Fixed a bug in
rtmvt that returned incorrect output (tranposed mean vector)
TruncatedNormal 2.1 (2019-08-31)
Bug fixes
- Cholesky decomposition with permutation now checks the arguments to ensure
lb less than ub, to avoid segfault errors. In case of degenerate bounds, the conditional distribution is used with fixed components for the degenerate variables (issue #2)
Changes
- Moved back probit example in documentation
- Change back to
randtoolbox package following its reuploading on CRAN
TruncatedNormal 2.0 (2019-06-10)
Bug fixes
- Throw error if
NaN in bounds
- Handle univariate cases in
mvrandt and mvrandn
Changes
- Can specify mean vector directly in the functions
- Functions for multivariate Student simulation ported from Matlab
- Change to the interface
- Many functions now internal (documented)
- Cholesky matrix permutation (GGB and GE reordering now supported)
- Selected code rewritten in Rcpp
- New vignette
- Removed random seed
- Change to qrng package for QRN generation of the Sobol sequence (
randtoolbox package unmaintained), but the sequence is not scrambled
- Use
nleq solver to solve convex program (also check KKT conditions).
TruncatedNormal 1.0 (2015-11-10)