Timeline for Diagnostic plot for assessing homogeneity of variance-covariance matrices
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 2, 2020 at 21:54 | vote | accept | Scortchi♦ | ||
| Jun 2, 2020 at 17:48 | comment | added | whuber♦ | @Gung The pooled ellipse is a reference. It deserves the same kind of visual prominence as any reference, such as the box of a boxplot. Although displaying blocks of a large matrix is what one often must resort to, it is less than satisfactory because it ruins the gestalt of seeing the entire thing. | |
| Jun 2, 2020 at 16:54 | comment | added | gung - Reinstate Monica | Why is it considered important for the pooled elipse to be more prominent? To me that seems to make it harder to see the distinctions amongst the rest, especially in the smaller panels w/ a plot matrix. (However, I haven't seen it the other way, it may well be worse for different reasons.) Regarding the issue of too many variables to see in a plot matrix, one possibility is to break out sections of the matrix. I did that in my answer to How to extract information from a scatterplot matrix when you have large N, discrete data, & many variables? | |
| Jun 2, 2020 at 16:30 | history | answered | whuber♦ | CC BY-SA 4.0 |