lagr.knn: Calculate the proportion of observations that lie within a...

In wrbrooks/lagr: Local adaptive grouped regularization

Description

Facilitates a knn-type bandwidth by calculating the proportion sum(w) / n. Called repeatedly by optimize in lagr.dispatch to home in on the desired knn.

Usage

lagr.knn(bw, loc, coords, dist, kernel, target, prior.weights, total.weight,
  verbose)

Arguments

bw	bandwidth for which to compute sum(w)
loc	location around which to center the kernel
coords	matrix of observation locations
dist	vector of distances from central location to the observation locations
kernel	kernel function for generating the local observation weights
target	targeted knn bandwidth
prior.weights	vector of prior observation weights provided by the user
total.weight	sum of prior weights
verbose	print detailed information about our progress?

Value

difference between the calculated sum(w) / total.weight and the target

wrbrooks/lagr documentation built on May 4, 2019, 11:59 a.m.