The space time separation is a broadly-used method of detecting non-stationarity
and temporal correlations in the time series being analyzed. The space time
separation plot is also used to select a proper Theiler window by selecting
a temporal separation enough to saturate the contour lines.
Usage
spaceTimePlot(takens = NULL, time.series = NULL, embedding.dim = 2,
time.lag = 1, max.radius = NULL, time.step = 1,
number.time.steps = NULL, numberPercentages = 10, do.plot = TRUE, ...)
## S3 method for class 'spaceTimePlot'
contourLines(x)
## S3 method for class 'spaceTimePlot'
plot(x, main = "Space time separation plot",
xlab = NULL, ylab = NULL, type = "l", ylim = NULL, col = NULL,
pch = NULL, add.legend = TRUE, ...)
Arguments
takens
Instead of specifying the time.series, the embedding.dim and the time.lag, the user
may specify directly the Takens' vectors.
time.series
The original time series being analyzed.
embedding.dim
Integer denoting the dimension in which we shall embed the time series.
time.lag
Integer denoting the number of time steps that will be use to construct the
Takens' vectors.
max.radius
Maximum neighbourhood radius in which the algorithm will look for finding neighbours. This
parameter may be used to avoid heavy computations. If the user does not specify a radius, the algorithm estimates it.
time.step
Integer denoting the number of discrete steps between two calculations of the space time plot.
number.time.steps
Integer denoting the number of temporal jumps in steps of time.step in which we want to compute the space time plot.
numberPercentages
Number of contour lines to be computed. Each contour line represent a concrete percentage of points (see Details).
do.plot
Logical. If TRUE, the time space plot is shown.
...
Additional plotting parameters.
x
A spaceTimePlot object.
main
A title for the plot.
xlab
A title for the x axis.
ylab
A title for the y axis.
type
Type of plot (see plot).
ylim
Numeric vector of length 2, giving the y coordinates range.
col
Vector of colors for each of the percentages of the plot.
pch
Vector of symbols for each of the percentages of the plot.
add.legend
add a legend to the plot?
Details
Each contour line of the space time plot indicate the distance you have to go (y-axis) to find a given fraction of
neighbour pairs, depending on their temporal separation (x-axis).
WARNING: The parameter number.time.steps should be used with caution since this
method performs heavy computations.
Value
A timeSpacePlot object that consist, essentially, of a matrix storing the values for each contour line.
Each row stores the value for a given percentage of points. Each column stores the value of the radius you have to go to
find a given fraction of neighbour pairs (the rows), depending on their temporal separation (the colums). This matrix
can be accessed by using the contourlines method.
The contourLines function returns the contour lines of the
space time plot.
Author(s)
Constantino A. Garcia
References
H. Kantz and T. Schreiber: Nonlinear Time series Analysis (Cambridge university press)
Examples
## Not run:
tak = buildTakens(sin(2*pi*0.005*(0:5000)),2,1)
stp.test = spaceTimePlot(takens=tak,number.time.steps=400,do.plot=TRUE)
## End(Not run)