We derive consistent and asymptotically normal estimators for the drift and volatility parameters of the stochastic heat equation driven by an additive space-only white noise when the solution is sampled discretely in the physical domain. We consider both the full space and the bounded domain. We establish the exact spatial regularity of the solution, which in turn, using power-variation arguments, allows building the desired estimators. We show that naive approximations of the derivatives appearing in the power-variation based estimators may create nontrivial biases, which we compute explicitly. The proofs are rooted in Malliavin-Stein's method.
翻译:当溶液在物理域内被分解地取样时,我们得出对由添加式空间专用白噪音驱动的随机热方程式的漂移和波动参数的一致和零星的正常估计值。我们既考虑整个空间,又考虑封闭域。我们确定溶液的准确空间规律性,这反过来,利用动力变换参数,又允许建立理想的估测器。我们显示,在以动力变换为基础的估测器中出现的衍生物的天真近似可能会产生非三角偏差,我们明确地计算了这些偏差。证据植根于Malliavin-Stein的方法。