Dear All The recently released version of the sampleSelection package [1] has the version number 1.0, because we consider the package now to be mature and feature-complete. Compared to version 0.6 of the sampleSelection package (which is described in an article in the Journal of Statistical Software [2]), the new version has, for instance, following new features: * estimation of binary-outcome selection models * probit models, tobit-2 models, and binary-outcome selection models can now be estimated with observation-specific weights * predict() methods for calculating unconditional and conditional expectations * improved/extended residuals() methods * logLik() methods * nobs() and nObs() methods * default df.residual() method works * removed the deprecated function tobit2() Any kind of feedback is--as always--very welcome, preferable via R-Forge [3]! Best regards, Ott and Arne [1] http://cran.r-project.org/package=sampleSelection [2] http://www.jstatsoft.org/v27/i07/ [3] http://r-forge.r-project.org/projects/sampleselection/ -- Arne Henningsen http://www.arne-henningsen.name