DecomposeR (version 1.0.6)

simple.ssa: Simple SSA decomposition

Description

Simple wrapper for Singular Spectrum Analysis, using the functions of the Rssa package (which is not installed by default by the DecomposeR package, you should install it independently). This function allows unevenly sampled data.

Usage

simple.ssa(xy, dt, n = 10, remove = "trend", groups = list(), plot = T, ...)

Value

a list made of $xy (original signal), $dt (depth/time), $m (a matrix of the decomposition), $repl (the replicate id of each point) and $mode (the mode id of each point).

Arguments

xy

signal to be decomposed

dt

depth/time

n

maximum amount of components

remove

whether to remove a linear trend ("trend", is the default), a mean value ("mean"), or to decompose as is (any other value)

groups

which components to regroup (list of the indices of elementary components to be regrouped, the entries of the list can be named, see the reconstruct() function in the Rssa package for more information)

plot

whether to show a visualisation of the importance of each component

...

any arguments to by given to the ssa() function (see Rssa package for more information)

Examples

set.seed(42)

n <- 600
t <- seq_len(n)

p1 <- 30
p2 <- 240

xy <- (1 + 0.6 * sin(t*2*pi/p2)) * sin(t*2*pi/p1)  + 2 * sin(t*2*pi/p2) +
  rnorm(n, sd = 0.5) + 0.01 * t

inter_dt <- round(runif(length(xy), min = 0.5, max = 1.5),1)

dt <- cumsum(inter_dt)

res <- simple.ssa(xy, dt, groups = list(c(1,2), c= 3:10))

parsimony(res)

integrity(xy, res)

if (FALSE) {
plot_emd(res, style = 1)}