spline_score: Univariate score estimation via the smoothing spline method...

In drape: Doubly Robust Average Partial Effects

Univariate score estimation via the smoothing spline method of Cox 1985 and Ng 1994.

Description

Univariate score estimation via the smoothing spline method of Cox 1985 and Ng 1994.

Usage

spline_score(x, df = 5, tol = 0.001, nmax = NULL)

Arguments

x	vector of datapoints
df	vector of smoothing parameters for the non-parametric score estimator, corresponding to the effective degrees of freedom for a smoothing spline.
tol	numeric tolerance, minimum distance between neighbouring points, to avoid singularities.
nmax	if specified, overrides tol as maximal number of unique points.

Value

score function "rho" and derivative "drho", which take vector input and yield a vector of score estimates corresponding to each df (in a list if there are multiple df values). Also output the vector "df".

Examples

# Single bandwidth
x <- stats::rlogis(100)
spl <- spline_score(x, df=6)
spl$rho(x)
spl$drho(x)

# Multiple bandwidths simultaneously
x <- stats::rt(n=100, df=4)
spl <- spline_score(x, df=c(2,5,10))
spl$rho(x)

drape documentation built on April 3, 2025, 9:23 p.m.