A total of 170 geological characteristics are used to describe each of the 58 deposits of a control sample which represents the population of commercial porphyry-copper-molybdenum deposits of the Cordilleran Belt of North and South America. The data are coded according to a binary system denoting presence or absence, and are assembled into a data matrix. The data matrix is processed by the technique of characteristic analysis into an incidence-coincidence matrix from which "typicality" loading factors are derived for the ordinal ranking of the 170 characteristics. The 44 most typical characteristics (upper quartile) are retained for the formulation of geological and geostatistical models.A composite geological model of the most typical commercialy porphyry deposit in the Cordilleran Belt is assembled for exploration guidance. Regression and classification models are formulated to correlate quantitatively the most typical geological characteristics to the gross dollar values of the porphyry deposits. An index is derived from the combination of the two statistical models and is used to rate objectively porphyry prospects of the Cordilleran Belt. The application of a similar method to complex geological targets other than porphyry deposits is suggested.