Abstract: A method for predicting the properties of crude oils or their boiling fractions which comprises selecting a chemical or perceptual or physical or performance property or groups of properties of the crude oil or its boiling fractions and creating a training set from reference samples which contain characteristic molecular species present in the crude oil or its boiling fractions. The reference samples are subjected to GC/MS analysis wherein the often collinear data generated is treated by multivariate correlation methods. The training set produces coefficients which are multiplied by the matrix generated from a GC/MS analysis of an unknown sample to produce a predicted value of the chemical, performance, perceptual or physical property or groups of properties selected.