A stochastic discrete choice model and its related estimation method are presented which allow to disentangle non-linear externalities from the intrinsic features of the objects of choice and from the idiosyncratic preferences of agents. Having veried for the ergodicity of the underlying stochastic process, parameter estimates are obtained through numerical methods and so is their statistical signicance. In particular, optimization rests on successive parabolic interpolation. Finally, the model and its related estimation method are applied to the case of rm localization using Italian sectoral census data.
A numerical estimation method for discrete choice models with non-linear externalities
Fabio Vanni
2014-01-01
Abstract
A stochastic discrete choice model and its related estimation method are presented which allow to disentangle non-linear externalities from the intrinsic features of the objects of choice and from the idiosyncratic preferences of agents. Having veried for the ergodicity of the underlying stochastic process, parameter estimates are obtained through numerical methods and so is their statistical signicance. In particular, optimization rests on successive parabolic interpolation. Finally, the model and its related estimation method are applied to the case of rm localization using Italian sectoral census data.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.