fftcor {timsac}

Auto And/Or Cross Correlations via FFT

Description

Compute auto and/or cross covariances and correlations via FFT.

Usage

  fftcor(y, lag = NULL, isw = 4, plot = TRUE, lag_axis = TRUE)

Arguments

y

data of channel X and Y (data of channel Y is given for isw = 2 or 4 only).

lag

maximum lag. Default is 2 \sqrt{n}, where n is the length of the time series y.

isw

numerical flag giving the type of computation.

1 :	auto-correlation of X (one-channel)
2 :	auto-correlations of X and Y (two-channel)
4 :	auto- and cross- correlations of X and Y (two-channel)

plot

logical. If TRUE (default), cross-correlations are plotted.

lag_axis

logical. If TRUE (default) with plot=TRUE, x-axis is drawn.

Value

acov

auto-covariance.

ccov12

cross-covariance.

ccov21

cross-covariance.

acor

auto-correlation.

ccor12

cross-correlation.

ccor21

cross-correlation.

mean

mean.

References

H.Akaike and T.Nakagawa (1988) Statistical Analysis and Control of Dynamic Systems. Kluwer Academic publishers.

Examples

# Example 1
x <- rnorm(200)
y <- rnorm(200)
xy <- array(c(x,y), dim = c(200,2))
fftcor(xy, lag_axis = FALSE)

# Example 2
xorg <- rnorm(1003)
x <- matrix(0, nrow = 1000, ncol = 2)
x[, 1] <- xorg[1:1000]
x[, 2] <- xorg[4:1003] + 0.5*rnorm(1000)
fftcor(x, lag = 20)