Distribution dynamics: a spatial perspective

It is quite common in cross-sectional convergence analyses that data exhibit spatial dependence. Within the literature adopting distribution dynamics approach, authors typically opt for prefiltering. We follow an alternative route and propose a procedure based on estimate of mean function conditional density which we develop two-stage non-parametric estimator allows dependence estimated via spline correlation function. The finite sample performance this assessed Monte Carlo simulations. apply incorporates proposed to per capita personal income US states metropolitan statistical areas.