k-Nearest-Neighbor Estimation (k-NN)

In this method, you fix the number of samples, $k_T$, that you want to consider and then find the smallest volume, $V_T$, around the point $x$ that encloses exactly that many samples.