This is an R package for fast, numerically-stable Gauss-Hermite quadrature. It uses the Golub-Welch algorithm (implemented using a Rcpp/LAPACK interface in compiled code) to evaluate quadrature rules with 1000+ points with numerical stability and efficiency. It also includes convenience functions for vanilla Gauss-Hermite quadrature and adaptive Gauss-Hermite quadrature based on Laplace approximations (Liu and Pierce 1994, Biometrika). All methods are currently unidimensional.
The author’s website is http://www.awblocker.com