This Title All WIREs
How to cite this WIREs title:
WIREs Comp Stat

# Level set tree methods

Can't access this content? Tell your librarian.

Level set trees provide a tool for analyzing multivariate functions. Level set trees are particularly efficient in visualizing and presenting properties related to local maxima and local minima of functions. Level set trees can help statistical inference related to estimating probability density functions and regression functions, and they can be used in cluster analysis and function optimization, among other things. Level set trees open a new way to look at multivariate functions, which makes the detection and analysis of multivariate phenomena feasible, going beyond one‐ and two‐dimensional analysis. This article is categorized under: Statistical and Graphical Methods of Data Analysis > Analysis of High Dimensional Data Statistical and Graphical Methods of Data Analysis > Statistical Graphics and Visualization Statistical and Graphical Methods of Data Analysis > Nonparametric Methods
(a) A perspective plot; (b) a contour plot; (c) a tree plot; (d) a plot of a volume function; (e) a barycenter plot of the first coordinate; (f) a barycenter plot of the second coordinate
[ Normal View | Magnified View ]
(a) A volume function with two levels to determine clusters; (b) a scatter plot with the corresponding three clusters
[ Normal View | Magnified View ]
(a) A volume function with the level to determine clusters; (b) a scatter plot with the corresponding two clusters; (c) a volume function with a higher level to determine clusters; (d) a scatter plot with the corresponding three clusters
[ Normal View | Magnified View ]
(a) A scatter plot of points, whose density is estimated in Figure ; (b) a colored scatter plot which groups those observations together, which are in the same branch of the level set tree
[ Normal View | Magnified View ]
(a) A tree plot; (b) a plot of a volume function; (c) a plot of the upper part of the volume function; (d) a barycenter plot of the first coordinate; (e) a barycenter plot of the second coordinate; (f) a barycenter plot of the third coordinate
[ Normal View | Magnified View ]
Contour plots of (a) projection ; (b) slice (x1, x2) ↦ f(x1, x2, 0); (c) projection ; (d) slice (x1, x3) ↦ f(x1, 0, x3); (e) projection ; (f) slice (x2, x3) ↦ f(0, x2, x3)
[ Normal View | Magnified View ]