In version 6.12, the module was used to compare prediction ellipses for robust and classical covariance matrices. A module called the CONTOUR module was in the version 6 (1989) documentation for SAS/IML. ![]() SAS/IML routines for computing prediction ellipses have been around for a long time. This can be useful for plotting ellipses for subgroups, ellipses that correspond to robust covariance estimates, or an ellipse for a population (rather than for a sample). Today's article describes the technique and shows how to use SAS/IML to compute a prediction ellipse from a 2 x 2 covariance matrix and a mean vector. ![]() The ELLIPSE statement draws the ellipse by using a standard technique that assumes the sample is bivariate normal. In a previous blog post, I showed how to overlay a prediction ellipse on a scatter plot in SAS by using the ELLIPSE statement in PROC SGPLOT.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |