Kernels can also be used as density estimators. Specifically, we have fh(x) = 1 n X i Kh(x − xi XXXXXXXXXXIn this setting we see again why it is important to have the integral of the kernel equal to...

Kernels can also be used as density estimators. Specifically, we have fh(x) = 1 n X i Kh(x − xi). (4.70) In this setting we see again why it is important to have the integral of the kernel equal to 1. Write a function kern_density that accepts a training vector x, bandwidth h, and test set x_new, returning the kernel density estimate from the Epanechnikov kernel. Visually test how this performs for some hand-constructed datasets and bandwidths.