bda (version 1.1.1-7)

histospline: Fit smoothed KDE to binned data.

Description

Fit smoothed KDE to binned data via an iterative algorithm.

Usage

histospline(x,f,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'.

gridsize

The size of grid where the density will be 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.

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)
x0 = seq(min(x),max(x),length=100)
plot(dmixnorm(x0,p,mu,s)~x0,type='l')
lines(density(x), col=3,lty=3)
lines(density(y), col=4,lty=4)
out = histospline(y)
lines(out,col=2,lty=2)