Learn R
Rssa (version 1.0.5)

plot: Plot SSA object

Description

This function plots various sorts of figures related to the SSA method.

Usage

# S3 method for ssa
plot(x,
     type = c("values", "vectors", "paired", "series", "wcor"),
     ...,
     vectors = c("eigen", "factor"),
     plot.contrib = TRUE,
     numvalues = nsigma(x),
     numvectors = min(nsigma(x), 10),
     idx = 1:numvectors,
     idy,
     groups)

Arguments

x

SSA object holding the decomposition

type

Type of the plot (see 'Details' for more information)

...

Arguments to be passed to methods, such as graphical parameters

vectors

For type = 'vectors', choose the vectors to plot

plot.contrib

logical. If 'TRUE' (the default), the contribution of the component to the total variance is plotted. For `ossa' class, Frobenius orthogonality checking of elementary matrices is performed. If not all matrices are orthogonal, corresponding warning is risen

numvalues

Number of eigenvalues to plot (for type = 'values')

numvectors

Total number of eigenvectors to plot (for type = 'vectors')

idx

Indices of eigenvectors to plot (for type = 'vectors')

idy

Second set of indices of eigenvectors to plot (for type = 'paired')

groups

Grouping used for the decomposition (see reconstruct)

Details

This function is the single entry to various plots of SSA objects. Right now this includes:

values

plot the graph of the component norms.

vectors

plot the eigenvectors.

paired

plot the pairs of eigenvectors (useful for the detection of periodic components).

series

plot the reconstructed series.

wcor

plot the W-correlation matrix for the reconstructed objects.

Additional (non-standard) graphical parameters which can be transfered via ...:

plot.type

lattice plot type. This argument will be transfered as type argument to function panel.xyplot.

ref

logical. Whether to plot zero-level lines in series-plot, eigenvectors-plot and paired-plot. Zero-level isolines will be plotted for 2d-eigenvectors-plot.

symmetric

logical. Whether to use symmetric scales in series-plot, eigenvectors-plot and paired-plot.

useRaster

logical. For 2d-eigenvector-plot and wcor-plot, indicating whether raster representations should be used. 'TRUE' by default.

col

color vector for colorscale (for 2d- and wcor-plots), given by two or more colors, the first color corresponds to the minimal value, while the last one corresponds to the maximal value (will be interpolated by colorRamp)

zlim

for 2d-plot, range of displayed values

at

for 2d-eigenvectors-plot, a numeric vector giving breakpoints along the range of z, a list of such vectors or a character string. If a list is given, corresponding list element (with recycling) will be used for each plot panel. For character strings, values 'free' and 'same' are allowed: 'free' means special breakpoints' vectors (will be evaluated automatically, see description of cuts argument in 'Details') for each component. 'same' means one breakpoints' vector for all component (will be evaluated automatically too)

cuts

for 2d-reconstruction-plot, the number of levels the range of z would be divided into.

fill.color

color or 'NULL'. Defines background color for shaped 2d-eigenvectors plot. If 'NULL', standard white background will be used.

See Also

ssa-object, ssa plot.reconstruction,

Examples

Run this code
# \donttest{
# Decompose 'co2' series with default parameters
s <- ssa(co2)
# Plot the eigenvalues
plot(s, type = "values")
# Plot W-cor matrix for first 10 reconstructed components
plot(s, type = "wcor", groups = 1:10)
# Plot the paired plot for first 6 eigenvectors
plot(s, type = "paired", idx = 1:6)
# Plot eigenvectors for first 6 components
plot(s, type = "vectors", idx = 1:6)
# Plot the first 4 reconstructed components
plot(s, type = "series", groups = list(1:4))
# Plot the eigenvalues by points only
plot(s, type = "values", plot.type = "p")

# Artificial image for 2dSSA
mx <- outer(1:50, 1:50,
            function(i, j) sin(2*pi * i/17) * cos(2*pi * j/7) + exp(i/25 - j/20)) +
      rnorm(50^2, sd = 0.1)
# Decompose 'mx' with default parameters
s <- ssa(mx, kind = "2d-ssa")
# Plot the eigenvalues
plot(s, type = "values")
# Plot eigenvectors for first 6 components
plot(s, type = "vectors", idx = 1:6,
     ref = TRUE, at = "same", cuts = 50,
     plot.contrib = TRUE, symmetric = TRUE)
# Plot factor vectors for first 6 components
plot(s, type = "vectors", vectors = "factor", idx = 1:6,
     ref = TRUE, at = "same", cuts = 50,
     plot.contrib = TRUE, symmetric = TRUE)
# Plot wcor for first 12 components
plot(s, type = "wcor", groups = 1:12, grid = c(2, 6))

# 3D-SSA example (2D-MSSA)
data(Barbara)
ss <- ssa(Barbara, L = c(50, 50, 1))
plot(ss, type = "values")
plot(ss, type = "vectors", idx = 1:12, slice = list(k = 1),
     cuts = 50, plot.contrib = TRUE)
plot(ss, type = "vectors", idx = 1:12, slice = list(k = 1, i = 1))
plot(ss, type = "vectors", vectors = "factor", idx = 1:12, slice = list(k = 3),
     cuts = 50, plot.contrib = FALSE)
plot(ss, type = "series", groups = 1:12, slice = list(k = 1))
plot(ss, type = "series", groups = 1:12, slice = list(k = 1, i = 1))
plot(ss, plot.method = "xyplot", type = "series", groups = 1:12, slice = list(k = 1, i = 1))
# }

Run the code above in your browser using DataCamp Workspace