Rosenblatt-Parzen kernel density estimate

The baseline whose boundary failure motivates the whole book: a symmetric kernel near the edge of a bounded support leaks mass outside it, so the bias there is O(h) and does not vanish at the interior rate.

Usage

morie_fauzi_kde(x, grid = NULL, h = NULL)

Arguments

x

numeric sample.

grid

evaluation points; a data-driven grid if NULL.

h

bandwidth; Silverman's rule if NULL.

Value

list: grid, density, bandwidth, mass, interior_bias_order, boundary_bias_order, boundary_consistent (FALSE), n, method.

References

Fauzi and Maesono (2023), Ch. 1.

Examples

set.seed(1)
morie_fauzi_kde(stats::rexp(200), grid = c(0.5, 1, 2))$density
#> [1] 0.5847155 0.4453491 0.1581525