Automatic interaction detector (AID)

A method or the technique which uses a set of categorical explanatory variables to divide data into the groups which are relatively homogeneous with respect to the value of some of the continuous response variable of interest. At each stage, the splitting of a group into two parts is defined by one of the descriptive variables, a subset of its categories defining one of the parts and the left over categories the other part. Of the possible divisions, the one chosen is that which maximizes the between groups sum of the squares of response variable.

The groups eventually formed may frequently be useful in predicting the value of response variable for some future observation. Observe the classification and regression tree method or technique and chi-squared automated interaction detector.