Methods of characterizing similarity between measurements on entities, computer programs product and data carrier

Abstract

A method for characterizing the similarity between measurements on a plurality of entities comprising a first entity and a second entity comprises receiving measurements taken at a plurality of measurement points per entity. A model is defined comprising a stochastic process and a model function having values which depend on a set of parameters and the measurement points. A set (β) of parameters is estimated by fitting the model function to the measurements. Residual data is determined for at least a part of the plurality of measurement points for the first entity and the second entity by subtracting the fitted function from the measurements. A correlation coefficient for the first entity (i') and the second entity (i″) is estimated based on the determined residual data and the estimated correlation coefficient is used to characterize the similarity between the measurements. The model is defined such that the residual data is expected to have a deterministic component which depends on the measurement points and that dominates the estimate of the correlation coefficient. The correlation coefficient is estimated using an estimate for the entity average residue averaged over the measurement points of the first entity and using an estimate for the entity average residue averaged over the measurement points of the second entity.