plot.smacof | R Documentation |
These methods provide various 2D plots for SMACOF models.
## S3 method for class 'smacof'
plot(x, plot.type = "confplot", plot.dim = c(1,2), sphere = TRUE,
bubscale = 1, col = 1, label.conf = list(label = TRUE, pos = 3,
col = 1, cex = 0.8), hull.conf = list(hull = FALSE, col = 1,
lwd = 1, ind = NULL), shepard.x = NULL, identify = FALSE,
type = "p", pch = 20, cex = 0.5, asp = 1, main, xlab, ylab,
xlim, ylim, col.hist = NULL, ...)
## S3 method for class 'smacofR'
plot(x, plot.type = "confplot", what = c("both", "columns", "rows"),
plot.dim = c(1,2), col.rows = hcl(0), col.columns = hcl(240),
label.conf.rows = list(label = TRUE, pos = 3,
col = hcl(0, l = 50), cex = 0.8),
label.conf.columns = list(label = TRUE, pos = 3,
col = hcl(240, l = 50), cex = 0.8),
shepard.x = NULL, col.dhat = NULL, type = "p", pch = 20,
cex = 0.5, asp = 1, main, xlab, ylab, xlim, ylim, ...)
## S3 method for class 'smacofID'
plot(x, plot.type = "confplot", plot.dim = c(1,2), bubscale = 1,
col = 1, label.conf = list(label = TRUE, pos = 3, col = 1,
cex = 0.8), identify = FALSE, type = "p", pch = 20, cex = 0.5,
asp = 1, plot.array, main, xlab, ylab, xlim, ylim, ...)
x |
Object of class |
plot.type |
String indicating which type of plot to be produced: |
plot.dim |
Vector with dimensions to be plotted. |
main |
Plot title. |
xlab |
Label of x-axis. |
ylab |
Label of y-axis. |
xlim |
Scale x-axis. |
ylim |
Scale y-axis. |
type |
What type of plot should be drawn (see also |
pch |
Plot symbol. |
cex |
Symbol size. |
asp |
Aspect ratio. |
col |
Point color. |
sphere |
In case of spherical smacof, whether sphere should be plotted or not. |
bubscale |
Scaling factor (size) for the bubble plot. |
label.conf |
List with arguments for plotting the labels of the configurations in a configuration plot (logical value whether to plot labels or not, label position, label color). If |
hull.conf |
Option to add convex hulls to a configuration plot. Hull index needs to be provided. |
shepard.x |
Shepard plot only: original data (e.g. correlation matrix) can be provided for plotting on x-axis. |
identify |
If |
what |
For unfolding only: Whether row coordinates, column coordinates, or both should be plotted. |
col.rows |
Row colors in unfolding configuration plot. |
col.columns |
Column colors in unfolding configuration plot. |
col.dhat |
Shepard plot only: color specification of the dhats. For row conditional transformations in unfolding a vector of the length of the number of rows should be specified. |
label.conf.rows |
List with arguments for plotting the labels of the row configurations in an unfolding configuration plot (logical value whether to plot labels or not, label position, label color). |
label.conf.columns |
List with arguments for plotting the labels of the columns configurations in an unfolding configuration plot (logical value whether to plot labels or not, label position, label color). |
col.hist |
Color of the borders of the histogram. |
plot.array |
Array arrangements of plots for individual difference models (see details). |
... |
Further plot arguments passed: see |
mds()
and smacofSym()
create an object of class "smacof"
, unfolding()
, prefscal()
, and smacofRect()
produce "smacofR"
, and smacofIndDiff()
generates "smacofID"
.
Plot description:
- Configuration plot (plot.type = "confplot"
): Plots the MDS configuration.
- Residual plot (plot.type = "resplot"
): Plots the disparities (d-hats) distances against
the fitted distances.
- Shepard diagram (plot.type = "Shepard"
): Diagram with the observed dissimilarities against the fitted distances including (isotonic) regression line.
- Stress decomposition plot (plot.type = "stressplot"
): Plots the stress contribution in of each observation. Note that it rescales the stress-per-point (SPP) from the corresponding smacof function to percentages (sum is 100). The higher the contribution, the worse the fit.
- Bubble plot (plot.type = "bubbleplot"
, not available for rectangular SMACOF): Combines the configuration plot with the point stress contribution. The larger the bubbles, the worse the fit.
- Histogram (plot.type = "histogram"
: gives a weighted histogram of the dissimilarities. For optional arguments, see wtd.hist
.
For smacofIndDiff()
the residual plot, Shepard diagram, and stress plot are based on the sum of the residuals across individuals/ways. The configuration plot represents the group stimulus space (i.e., joint configuration). If plot.array
is not specified, it produces a Shepard plot of the distances summed across subjects, if plot.array = 0
it produces a sqrt(nsubjects) times sqrt(nsubjects) array of graph panels, if plot.array = 3
it produces 3x3 arrays of graph panels, if plot.array = c(2, 3)
it produces 2x3 arrays of graph panels, and if plot.array = c(3, 2, 5)
produces 3x2 arrays of panels (only the first two values are used).
plot.procr
## 2D plots for simple MDS
data(trading)
res <- mds(trading)
plot(res, plot.type = "confplot")
plot(res, plot.type = "confplot", label.conf = list(pos = 5)) ## avoid overlapping labels
plot(res, plot.type = "Shepard")
plot(res, plot.type = "stressplot")
plot(res, plot.type = "resplot")
plot(res, plot.type = "bubbleplot")
plot(res, plot.type = "histogram")
## Add convex hulls to configuration plot
r <- cor(PVQ40, use = "pairwise.complete.obs")
diss <- sim2diss(r, method = "corr")
res <- mds(delta = diss, type = "ordinal")
codes <- substring(colnames(PVQ40), 1, 2) ## supplementary variable
plot(res, hull.conf = list(hull = TRUE, ind = codes, col = "coral1", lwd = 2))
## Shepard plots
ekmanD <- sim2diss(ekman)
fit1 <- mds(ekmanD, type = "ordinal")
plot(fit1, plot.type = "Shepard")
plot(fit1, plot.type = "Shepard", shepard.x = ekman) ## original data on x-axis
## Joint configuration plot and row/column stressplots for unfolding
data(breakfast)
res <- unfolding(breakfast)
plot(res, plot.type = "confplot")
plot(res, plot.type = "stressplot")
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.