nlpden: Density estimation by quadratic programming

In swager/nlpden: Density estimation by quadratic programming

Description

Estimates the density of a signal that has been convolved with a known noise distribution. For example, if X = mu + epsilon where epsilon is noise drawn from a known noise_density_function, then nlpden will return an estimate of the density f(X).

Usage

nlpden(noisy_signal_raw, noise_density_fn = dnorm, BIN_SZ = 0.01,
  MARGIN = 1)

Arguments

noise_density_fn	the density of the additive noise function
BIN_SZ	The size of the histogram bins for the empirical distribution
MARGIN	The amount of padding on each side of the data
noisy_signal	the raw observations

Value

density estimate of the form (x, f(x))

Examples

mu = c(runif(1000, min = -3, max = 3), rep(0, 1000))
X = mu + rnorm(2000)
f.hat = nlpden(X)
plot(f.hat, type = "l")

swager/nlpden documentation built on May 30, 2019, 9:33 p.m.