Nonlinear estimation with associative memories and machine evaluation of derivatives: an application to calibrating spatial interaction models

In this paper we apply the theory of linear associative memoriesin producing initial parameter estimates for nonlinear iterative approaches.We also propose the use of FEED (Fast and Efficient Evaluation of Derivatives) to evaluate partial derivatives of functions encountered in nonlinearestimation. Suggested methods are presented in the context of calibratingspatial interaction models and are illustrated through numerical examples.