The most stable version of the package
gmm can be downloaded from
CRAN. The development version is available on
R-Forge.
The package has changed quite a bit since I published the vignette in the Journal of Statistical Software in 2010. See the NEWS file for details. I try to include as much as I can new developments in econometrics. For example, I recently added the method of bandwidth selection proposed by Wilhelm (2015) in Econometric Theory, for the computation of the HAC covariance matrix of the moment vector. It is also possible to estimate a model by GEL using the Hellinger Distance of Kitamura-Otsu-Evdokimov (2013) or the Exponentially tilted Hellinger Distance of Antoine-Dovonon (2015).
The package
momentfit will eventually replace the
gmm package. It is available on CRAN and the development version is available on
R-Forge. It is entirely built with S4-type objects and methods. Several classes for moment-based models are defined. The classes are defined for moment conditions derived from a single equation or a system of equations. The conditions can also be expressed as functions or formulas. Several methods are also offered to facilitate the development of different estimation techniques.
On the same R-Forge development page, there is a package called
gmmExtra. It is under development and far from being ready for CRAN. The idea of this package is to include related methods. You will find methods such as the one proposed by Kleibergen (2005) for testing hypotheses under weak identification. I have also implemented the bootstrap method of Inoue and Shintani (2006). The package does not build for Windows yet. You can install it if you are either a Mac or Linux user.