Hi all,
I am trying to perform an LDA analysis on >100 variables to separate 4 groups.
I am using the following:
discrm lda x y z, group (group)
However after running it I get the following error:
"pooled within-group SSCP matrix has rank 88 instead of rank 115"
When i reduce the number of variables to about 50% the analysis is run without issues.
Can anyone comment on this? Is it caused by the number of variables obtained?
Also, how can I display the generated LD functions? similar to PCA generated PC?
Thank you in advance for the help and time.
Giuseppe
I am trying to perform an LDA analysis on >100 variables to separate 4 groups.
I am using the following:
discrm lda x y z, group (group)
However after running it I get the following error:
"pooled within-group SSCP matrix has rank 88 instead of rank 115"
When i reduce the number of variables to about 50% the analysis is run without issues.
Can anyone comment on this? Is it caused by the number of variables obtained?
Also, how can I display the generated LD functions? similar to PCA generated PC?
Thank you in advance for the help and time.
Giuseppe
