rollfun

A collection of rolling functions.

Among others

Exponential moving average, Iterative exponential moving average, Optimized windowed convolution filter, Arbitrary rolling function, Iterative rolling mean, Rolling RMS, Windowed rolling RMS.

As well as

generate trapezoid windows and lot of other popular windows (epanechnikov, biweight, triweight, hann, hamming, blackmann, kaiser-bessel, poisson, lancsoz, ...)

library(devtools)
install_github("AkselA/R-rollfun")
library(rollfun)

# rolliter() and iema() works well on data with high rate of missingness
set.seed(1)
x <- rnorm(150, 0, 20) + c(1:50, rep(50, 50), 50:1)
x2 <- x
x2[sample(2:149, 50)] <- NA

plot(x, col="lightgrey")
points(x2, pch=16, cex=0.3)

lines(iema(x2, 100, dir="av"), col="red")
lines(rolliter(x2, 6, 10, partial=TRUE), col="blue")

To-do

• NA-enable ema()
• zoo methods