Probabilistic interpretation and numerical approximation of a Kac equation without cutoff (Q1613657)

The noncutoff Kac equation \(\partial f/\partial t= K_{\beta }(f,f)\) is considered, where \(f=f(v,t), t\geq 0, v\in \mathbb{R}\), and \[ K_{\beta }(f,f)(t,v)= \int _{v_*\in R}\int _{\theta = -\pi }^{\pi } (f(t,v')f(t,v'_*)-f(t,v)f(t,v_*))\beta (\theta) d\theta dv_* \] with \(v'=v\cos \theta -v_*\sin \theta\), \(v_*'=v\sin \theta +v_*\cos \theta\), where the cross-section \(\beta :[-\pi , \pi ]\setminus \{ 0\} \to \mathbb{R}_+\) is an even function satisfying \(\int _0^{\pi } \theta ^2\beta (\theta) d\theta <\infty \) in the cutoff case. A nonlinear pure-jump Markov process is associated with this Kac equation, which is shown to be the unique solution of a nonclassical stochastic differential equation. Its law is approximated by simulable stochastic particle systems, and the precise rate of convergence is given. The authors also perform an empirical study of convergence.