Let (θ1,X1),…, (θn,Xn), (θ, X) be iid random vectors ,where θ∈{0,1},X∈Rd Denote by θ′n the nearest neighbour discriminator of θ based on the training samples (θ1,X1),…, (θn,Xn) and the observed X; put(?). This paper gives a sufficient and necessary condition for (?) as n→∞, namely (P(θ=0, X=x)-P(θ=1, X=x))2·P(θ=0, X=x)·P(θ=1, X=x)=0 for every x∈Rd.This generalizes a previous result of the authors [5] and improves a result of Wagner, T.J. [2]. |