Abstract

A set of three statistical tests can be applied to all geophysical modelling techniques to objectively determine the 'goodness of fit' between the observed and modelled data. The three tests examine the mean, the standard error and the number of runs in the data defining the error function and they accept or reject the model on the basis of standard statistical tests on these three measures. The application of these tests to automatic, interactive or user controlled modelling procedures provides an objective and consistent set of criteria for accepting or rejecting a model on statistical grounds.

You do not currently have access to this article.