The problem considered here is the estimation of the probability density function f (x1, ⋯, xp) at a point z = (z1, ⋯, zp) where f is positive and continuous. An estimator is proposed and consistency ...