STPGA: Selection of Training Populations by Genetic Algorithm
Combining Predictive Analytics and Experimental Design to Optimize Results. To be utilized to select a test data calibrated training population in high dimensional prediction problems and assumes that the explanatory variables are observed for all of the individuals. Once a "good" training set is identified, the response variable can be obtained only for this set to build a model for predicting the response in the test set. The algorithms in the package can be tweaked to solve some other subset selection problems.
||R (≥ 2.10), AlgDesign, scales, scatterplot3d, emoa, grDevices
||R.rsp, EMMREML, quadprog, UsingR, glmnet, leaps, Matrix
||Deniz Akdemir <deniz.akdemir.work at gmail.com>
Please use the canonical form
to link to this page.