LAUSR

Abstract System identification is a data-driven input-output modeling approach more and more used in biology and biomedicine. In this application context, several assays are repeated to estimate the response variability and reproducibility. The inference of the modeling conclusions to the whole population requires to account for the inter-individual variability within the modeling procedure. One solution consists of using mixed effects models but up to now no similar approach exists in the system identification literature. In this article, we propose a first solution based on an ARX (Auto Regressive model with eXternal inputs) structure using the EM (Expectation-Maximisation) algorithm for the estimation of the model parameters. Using the Fisher information matrix, the parameter standard errors are estimated; this allows for group comparison tests. Simulations show the relevance of this solution compared with a classical procedure of system identification repeated on each subject. Taking into account all the information available in the population allows to gather the parameters between individuals.