You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Having NA or a constant value on the diagonal is not ideal because the desired information here is the sample variance, which is specific to each variable. Instead, users expect a complete sample covariance matrix, similar to the output obtained using:
I know
colpair_map
is a solution in relation to #42. In the case ofcolpair_map
we have the following forcor
:Created on 2024-02-27 with reprex v2.1.0
While this approach makes sense for the sample correlation, it's not suitable for the sample covariance (
cov
).Created on 2024-02-27 with reprex v2.1.0
Having
NA
or a constant value on the diagonal is not ideal because the desired information here is the sample variance, which is specific to each variable. Instead, users expect a complete sample covariance matrix, similar to the output obtained using:Created on 2024-02-27 with reprex v2.1.0
It will be possible to display the diagonal elements of the covariance matrix by adding an option to the
.diagonal
argument?The text was updated successfully, but these errors were encountered: