A normal contour ellipsoid is a 3-dimensional ellipse that encompasses a specified portion of points. The ellipsoid is computed from a contour of the multivariate normal distribution fit to the points. The ellipsoid is a function of the means, standard deviations, and correlations of variables on the plot. See the Multivariate Methods book for details about multivariate normal distributions.
•
|
Coverage changes the portion of data points covered by the ellipsoid. The larger the value, the bigger the ellipsoid.
|
•
|
Transparency changes the surface of the ellipsoid from transparent to opaque. The larger the value, the more opaque the ellipsoid.
|
You display and remove normal contour ellipsoids by selecting and deselecting Normal Contour Ellipsoids from the red triangle menu.
The examples in this section use the Iris.jmp sample data table, which includes measurements of sepal length, sepal width, petal length, and petal width for three species of iris.
This feature is particularly valuable when you have many points on a 3D scatterplot; the contours can be so dark that you cannot see the structure. In this situation, you remove the points so that only the contours are displayed. See Optimizing a Dense Nonparametric Density Contour for details.
You display and remove nonparametric density contours by selecting and deselecting Nonpar Density Contours from the red triangle menu.
|
|||||||
Type a new bandwidth for each variable, or click and drag the sliders. Click Apply to display your changes.
|
To remove points from a 3D scatterplot, select Show Points from the red triangle menu. You can further optimize the contours by changing their size, color, and transparency. See Descriptions of the Scatterplot 3D Options for details.
To customize properties such as the marker size, text size, and grid lines, right-click the 3D scatterplot and select Settings. The Settings window appears. As you modify the settings, a preview appears on the 3D scatterplot.