Insightful has released a Missing Data Library in Splus 6.0 on Unix and Windows platforms.

The library has code that fits the multivariate Gaussian, loglinear, and conditional Gaussian models using the EM and data augmentation (DA) algorithms. The DA algorithms also produce proper multiple imputations. The code is based on Joe Schafers' work on DA (and previous work by Rubin et. al. on EM and multiple imputation). Schafer was a consultant and subcontractor on this work which was partially funded through a National Institutes of Health SBIR grant.

The library builds upon Schafer's code, but in some cases uses different algorithms. For example, the EM to fit the Gaussian model uses a Cholesky decomposition of the covariance rather than sweeps as in Schafer's NORM function. Performing multiple complete data analyses after multiple imputation, and consolidating results, is simplified by using the library.

The library includes code, help files, and manual.