Cross-Validation

Cross-validation omits a point (red point) and calculates the value of this location using the remaining 9 points (blue points). The predicted and actual values at the location of the omitted point are compared. This procedure is repeated for a second point

In statistics, Cross-Validation is is a model validation technique for assessing how the results of a statistical analysis will generalize to an independent data set. It is mainly used in settings where the goal is in prediction, and how accurately a predictive model will perform in practice. It involves removing a control point from the data to be interpolated, using other control points to estimate a value at the location of the removed point, and then computing the residual, the difference between the known and estimated control point in turn. If cross-validation is done for a variety of interpolation methods, the resulting sets of residuals can be compared.

Cross-validation in ArcGIS

Cross-validation can be run in ArcGIS using the Cross-Validation (GeoStatistical Analysis) tool under Geostatistical Analysis Tool> Utilities>Cross Validation