The package cmdsr provides functions to compute and analyze continuous embeddings. Continuous embeddings are an extension of Multi-Dimensional Scaling (MDS). They are the solution of multiple MDS problems that are solved simultaneously together with a smoothing penalty. This ensures that the individual MDS solutions are similar local minima, so that they can be connected smoothly. More information about each function can be found in the function documentation.
Computing continuous embeddings
The function cmds
computes the continuous embedding. Its only obligatory parameter is the list of distance matrices DL
. The most commonly used optional parameters are k
, the dimension of the MDS embedding and l
, the smoothing parameter.
Plotting continuous embeddings
The function plot.cmds
can be used to plot the results from cmds
. plot.cmds(res)
plots the continuous embedding. It automatically detects the embedding dimension. Setting the optional parameter shepard
to TRUE
yields a Shepard plot, a common diagnostic tool for MDS. A plot of the convergence trace is computed when calling plot.cmds
with convergence = TRUE
. Two plots are created. The second is similar to the first, but skips the first iteration. This results in better scaling of the y-axis.
Summarizing continuous embeddings
The function summary.cmds
gives a summary of the cmds
results.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.