Basis Selection from Multiple Libraries BSML stands for Basis Selection from Multiple Libraries and it is a methodology for spatially-adaptive multivariate regression. To fit a nonparametric or semiparametric regression model, multiple libraries may be used to approximate the mean function. The program will adaptively select basis functions from each library and find the optimal model complexity by using the Generalized Degrees of Freedom or Covariance Penalty. Software Depends: R (>= 2.5.1) Suggests: assist, nlme, leaps Junqing Wu <wjqpu@yahoo.com> Junqing Wu <wjqpu@yahoo.com>, Jeff Sklar <jsklar@calpoly.edu>, Yuedong Wang <yuedong@pstat.ucsb.edu>, Wendy Meiring <meiring@pstat.ucsb.edu> GPL (>= 2) 2012-04-23 application/tgz http://CRAN.R-project.org/package=bsml