NEWS

---

Spbsampling 1.3.5 (2022-08-24)

• Added JSS reference.

Spbsampling 1.3.4 (2020-08-24)

• Functions pwd() and swd() now return a list with the following components: s, the samples stored in a matrix nrepl * n, and iterations, the number of iterations run by the algorithm.
• Functions stprod() and stsum() now return a list with the following components: mat, the standardized matrix, iterations, the number of iterations run by the algorithm, and conv, the convergence reached by the algorithm.

Spbsampling 1.3.3 (2020-03-31)

• Fixed typos in some examples.

Spbsampling 1.3.2 (2020-03-04)

Usage

• In the functions pwd(), swd() and hpwd() changed the name of the parameter nsamp in n.

Under the hood

• In the functions pwd() and swd() introduced stopping criterion.

Spbsampling 1.3.1 (2019-12-16)

Usage

• In the functions pwd(), swd() and hpwd() changed the name of the parameter bexp in beta.
• In the functions stprod() and stsum() changed the name of the parameter vecin con.

Under the hood

• In the functions pwd() and swd() replaced arma::sort_index() with arma::stable_sort_index().
• In the functions pwd(), swd() and hpwd() fixed the sample size check.

Spbsampling 1.3.0 (2019-08-04)

New features

• In the functions pwd() and hpwd() a new parameter called bexp has been introduced, which regulates the amount of spread of the samples. Default value set to 10.
• In the function swd() the parameter bexp is now set to double and with default equal to 10.

Minor improvements

• Minor improvements of guides:
  • some updates to the examples;
  • removed some non-ASCII characters;
  • inserted links to reference websites.

Spbsampling 1.2.0 (2019-03-12)

• Changed output of functions pwd(), swd() and hpwd() from a matrix of dimension (nsamp * nrepl) * 2 to a matrix of dimension nrepl * nsamp.
• Removed Rdpack from Imports.
• Minor improvements of guides.
• Changed name of file heurprod.cpp to hpwd.cpp

Spbsampling 1.1.0 (2019-01-21)

New features

• Added a new function sbi() for compute the spatial index of a sample, implemented in C++ through the Armadillo library, using Rcpp and RcppArmadillo.
• In the functions pwd() and swd() the parameter nrepl is now set to default at 1.
• In the functions pwd() and swd() the parameter niter is now set to default at 10.
• In the functions stprod() and stsum() the parameter differ is now set to default at 1e-15.
• In the functions stprod() and stsum() the parameter niter is now set to default at 1000.
• In the functions stprod() and stsum() there is no print for each convergence step anymore.

Improvements under the hood

• The functions pwd() and swd() now take advantages of the Armadillo library.
• The function stprod() is now implemented in C++ through the Armadillo library, using Rcpp and RcppArmadillo.
• The function stsum() is now implemented in C++ through the Armadillo library, using Rcpp and RcppArmadillo.
• The function hpwd() is now implemented in C++ through the Armadillo library, using Rcpp and RcppArmadillo.
• In all the functions there are now checks about the correct inputs.

Minor improvements

• Correction of some typos along the guides.

Miscellaneous

• Added a NEWS.md file to track changes to the package.
• Updated DESCRIPTION file.

Spbsampling 1.0.0 (2018-05-30)

• First release on CRAN.