bda (version 3.1.3-2)

smkde: Fit smoothed KDE to binned data.

Description

Fit smoothed KDE to binned data via an iterative algorithm.

Usage

smkde(x,f,bw,gridsize=512L,na.rm=TRUE,just="center",binned=FALSE, scale=1.0)

Arguments

x

vector of raw data or distinct center information of binned data.

f

A vector of frequencies. Can not be missing if 'binned=TRUE'.

bw

Smooth parameter.

gridsize

The size of grid points where the density willbe evaluated.

na.rm

Default "TRUE". Automatically remove missing values. Otherwise, stop and warn.

just

Character showing how data are binned. "center", "left" or "right".

binned

Logic value. Default value: 'FALSE'.

scale

Numeric. Default=1.0.

References

Blower, Gordon and Kelsall, Julia, (2002). Nonlinear kernel density estimation for binned data: convergence in entropy. Bernoulli, 8(4), 423-449.

See Also

bfmm.

Examples

p = c(.4,.6)
mu = c(34.5,40)
s=c(1.5,1.3)
x = rmixnorm(300,p,mu,s)
y = round(x,0)
out = smkde(y)