sliding_window: Compute summary statistics on a sliding window
In DominikMueller64/LDtools: Tools for Linkage Disequilibrium Analysis

Description Usage Arguments Details Value Author(s) Examples

Compute summary statistics on a sliding window along a vector containing data where the positions of the data points are stored in another vector.

1 2	sliding_window(x, pos, start, width, advance, stat = c("mean", "median", "min", "max", "sd"))

`x`	A numeric vector containing the values.
`pos`	A incresingly sorted vector containing the corresponding positions.
`start`	A double. The starting point of the sliding window.
`width`	A double. The width of the sliding window.
`advance`	A double. The step size of the sliding window.
`stat`	A string. The summary statistic to be calculated.

Note that only windows that fully fit into the range of pos are considered.

A data.frame with columns start (start of the window), end (end of the window), stat (the computed summary statistic) and n (number of data points in the window).

Dominik Mueller (dominikmueller64@yahoo.de).

set.seed(123L)
n <- 1000L
x <- arima.sim(n = n, list(ar = 0.99))
pos <- sort(runif(n))
plot(x = pos, y = x)
advance <- 0.01
width <- c(0.02, 0.1, 0.2)
colors <- c('red', 'green', 'blue')
for (i in seq_along(width)) {
  df <- sliding_window(x, pos, 0.0, width[i], advance, "mean")
  points(x = df$begin + width[i] / 2, df$stat, col = colors[i], pch = 19, lwd = 2.0, type = 'l')
}