The squared L2 norm of a probability density function derivative measures roughness and appears in optimal smoothing constants. For , its directional derivative along a smooth compactly supported perturbation is . When integration by parts is valid twice, this equals . Thus regular root- estimation requires substantially more regularity than merely estimating a smooth probability density function.
Let be an independent and identically distributed sample from probability density function , take and , and write . Assume the kernel for density estimation has square-integrable second derivative. DefineThis U-statistic obeys whenever differentiation and the displayed integrals are justified. Under the usual convolution regularity this is .
The exact relation to the second derivative kernel density estimator isIndeed, expand the square into ordered pairs; its diagonal terms are all . Both subtracting this diagonal and changing the normalization from to are needed for the displayed U-statistic. This removes the diagonal sampling contribution, while retaining smoothing bias of an estimator relative to . The corrected value can be negative in a finite sample even though its target is nonnegative.
Articles by others on the same topic
There are currently no matching articles.