plot.MclustDR.Rd
Graphs data projected onto the estimated subspace for modelbased clustering and classification.
# S3 method for MclustDR plot(x, dimens, what = c("scatterplot", "pairs", "contour", "classification", "boundaries", "density", "evalues"), symbols, colors, col.contour = gray(0.7), col.sep = grey(0.4), ngrid = 200, nlevels = 5, asp = NULL, ...)
x  An object of class 

dimens  A vector of integers giving the dimensions of the desired coordinate projections for multivariate data. 
what  The type of graph requested:

symbols  Either an integer or character vector assigning a plotting symbol to each
unique mixture component. Elements in 
colors  Either an integer or character vector assigning a color to each
unique cluster or known class. Elements in 
col.contour  The color of contours in case 
col.sep  The color of classification boundaries in case 
ngrid  An integer specifying the number of grid points to use in evaluating the classification regions. 
nlevels  The number of levels to use in case 
asp  For scatterplots the \(y/x\) aspect ratio, see

...  further arguments passed to or from other methods. 
Scrucca, L. (2010) Dimension reduction for modelbased clustering. Statistics and Computing, 20(4), pp. 471484.
Luca Scrucca
if (FALSE) { mod < Mclust(iris[,1:4], G = 3) dr < MclustDR(mod) plot(dr, what = "evalues") plot(dr, what = "pairs") plot(dr, what = "scatterplot", dimens = c(1,3)) plot(dr, what = "contour") plot(dr, what = "classification", ngrid = 200) plot(dr, what = "boundaries", ngrid = 200) plot(dr, what = "density") plot(dr, what = "density", dimens = 2) data(banknote) da < MclustDA(banknote[,2:7], banknote$Status, G = 1:3) dr < MclustDR(da) plot(dr, what = "evalues") plot(dr, what = "pairs") plot(dr, what = "contour") plot(dr, what = "contour", dimens = c(1,3)) plot(dr, what = "classification", ngrid = 200) plot(dr, what = "boundaries", ngrid = 200) plot(dr, what = "density") plot(dr, what = "density", dimens = 2) }