Application of Non-linear Error Correction Models to the demand for Money in Tunisia

Non linear error correction models (NLECM) have been increasingly used recently in the econometric literature since they lead to more insightful results than the linear models in empirical studies. In this research we present alternative methods of dealing with non linearity in error correction models. We apply these models to the demand for money in Tunisia and our results show that we get better predictions than the usual linear models. It is also shown that the speed of adjustment towards the long-run equilibrium is faster than what we get using linear error correction models.