Module pointprocesses::estimators::kernels

Kernel structs.

Structs

EpanechnikovKernel	The Epanechnikov kernel is given by $$ K_h(x, x') = D\left( \frac{|x-x'|}{h} \right) $$ where $D(u) = \frac34(1-u^2)\mathbf{1}_{|u|\leq 1}$.
GaussianKernel	Homogeneous fixed-bandwidth Gaussian kernel, of the form $$ K_h(x, x') = \exp\left( -\frac{(x-x')^2}{2h^2} \right) $$
NearestNeighborKernel	Fixed-bandwidth nearest-neighbor (or uniform) kernel, $$ K_h(x, x') = \mathbf{1}_{|x - x'| < h} $$

Traits

RegKernel	Trait for non-parametric regression kernels of the form $$ K_h(x, x') = D\left(\frac{|x-x'|}{h}\right) $$
RegKernelMass	Trait for kernel mass integrals.