We propose a method to compute confidence bounds around the predictions from chi-squared automatic interaction detection (CHAID). These bounds are useful for ex ante target selection as some clusters may or may not overlap. They are also useful for ex post evaluation of the predictive quality of CHAID. The method is illustrated using simulated data.

CHAID, bootstraps, confidence intervals, target selection
dx.doi.org/10.1016/j.is.2005.03.002, hdl.handle.net/1765/13391
Information Systems: databases: their creation, management and utilization
Erasmus School of Economics

van Diepen, M, & Franses, Ph.H.B.F. (2006). Evaluating chi-squared automatic interaction detection. Information Systems: databases: their creation, management and utilization, 31(8), 814–831. doi:10.1016/j.is.2005.03.002