Genetic subsets regression

Subset regression procedures have been shown to provide better overall performance than stepwise regression procedures. However, due to the combinatorial nature of evaluating each potential subset, subset regression techniques are costly to use. To resolve this difficulty, the use of a simple genetic algorithm (GA) is proposed to reduce the number of subsets which must be evaluated. Any of a number of popular criteria, including Mallowsʹ Cp, MSE, R2, AIC, etc., can be used to drive the search strategy associated with the use of the GA. Several illustrated examples on its use are provided.