The Salford Predictive Modeler™ suite (SPM) includes a number of automated tools to assist in the process of feature selection under the BATTERY mechanism. For example,
BATTERY KEEP
Selects a subset of features at random and builds a model from this random subset only. The GUI will guide you in how to use this option, but from the command line you would issue something like:
BATTERY KEEP=100, 15
Which requests 100 models, each of which includes 15 randomly-selected predictors. If we are sure that we want certain variables included in every such model, the command would look like:
BATTERY KEEP=100, 15 CORE= X1, X2, X3, X4, X5
Which would pre-select five of the 15 available predictors, leaving another 10 to be selected at random.
BATTERY SHAVE is another automated model-building mechanism that is intended to start with a list of predictors that we would like to pare down to a smaller subset. We "shave" variables from the list of predictors by running a model, reviewing the variable and then rebuild the model. In another mode, we can remove the most important variables sequentially instead. Both modes can be very helpful in discovering a good set of predictors to work with.
BATTERY KEEP=‹NK, NR› [CORE=‹predictor›, ‹predictor›,…]
Repeat the model NR times, selecting a subset of NK predictors from the KEEP list each time.
The CORE option defines a group of predictors (from the main KEEP list) that are included in each of the models of the battery.
The CORE option, if used, must be the final option on the BATTERY command.<.p>
