Methods for Multivariate Quadrature

Provides methods to construct multivariate grids, which can be used for multivariate quadrature. This grids can be based on different quadrature rules like Newton-Cotes formulas (trapezoidal-, Simpson's- rule, ...) or Gauss quadrature (Gauss-Hermite, Gauss-Legendre, ...). For the construction of the multidimensional grid the product-rule or the combination- technique can be applied.


Changes in version 1.0-6 (2016-07-19) [new feature] The aggregation of the nodes and weights while the creation of a sparse grid is no longer done at the very end but repeatetly. This change saves a lot of memory - especially for high accuracy levels in high dimensions. [new feature] mvQuad do not longer depend on the "rgl"-package but suggests. This enables the use of mvQuad on systems without X11-environment (like HPC-Clusters)

Changes in version 1.0-5 (2016-02-18) [new feature] Nodes and weights for Gaussian quadrature (GLa, GLe, GHe) are now computed by "gauss.quad" (package: statmod). This enables use of higher order rules

Changes in version 1.0-4 (2015-11-11)

[bug fix] enable rescaling for 1D-grids

Changes in version 1.0-3 (2015-11-03)

[bug fix] aggregation of weights in 'createNIGrid' corrected (thanks to Daniel B.)

Changes in version 1.0-2 (2015-11-01)

[initial submission] editorial corrections

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


1.0-6 by Constantin Weiser, 5 years ago

Browse source code at

Authors: Constantin Weiser (HHU of Duesseldorf / Germany)

Documentation:   PDF Manual  

Task views: Numerical Mathematics

GPL-3 license

Imports data.table, statmod, methods

Suggests knitr, rgl, rmarkdown

Imported by AssetCorr, BLPestimatoR, ICAOD, aghq, bisque.

See at CRAN