Many non/semiparametric time series estimates may be regarded as different forms of sieve extremum estimates. For stationary absolute regular mixing observations, the authors obtain convergence rates of sieve extremurn estimates and root-n asymptotic normality of 'plug-in' sieve extremum estimates of smooth functionals. As applications to time series models, they give convergence rates for nonparametric ARX(p, q) regression via neural networks, splines, wavelets; root-n asymptotic normality for partial linear additive AR(p) models, and monotone transformation AR(1) models.

This is the first systematic experimental study of the comparative performance of two incentive‐compatible mechanisms for public goods provision: the basic quadratic mechanism by Groves and Ledyard and the paired‐difference mechanism by Walker. Our experiments demonstrate that the performance of the basic quadratic mechanism under a high punishment parameter is far better than that of the same mechanism under a low punishment parameter, which, in turn, is better than that of the paired‐difference mechanism. We estimate three individual behavioral models: an exponentialized relative payoff sum model outperforms the generalized fictitious play model. We also provide a sufficient condition for convergence under the basic quadratic mechanism.