Abstract
The aim of this paper is to analyse the behaviour of the Federal Reserve in the Greenspan era by using recently developed neuro-fuzzy techniques. Such models require the operation assumptions concerning the conduct of monetary policy to be set in the form of flexible rules. Moreover, the approach allows an adaptive model to be built, thus pointing out the role played by the evolution of the monetary policy decision mechanism. Besides the usual set of macro-economic variables, the input data set includes a stock market indicator. Our results show that the simulated series of federal funds rate mimics almost perfectly the actual behaviour of the monetary policy instrument.