2
$\begingroup$

When I plot a PCA the axis titles are PC1/2 - "% of explained var." and when I plot LDA the axis titles are LD1/2 - "% of explained between-group Variance". What would be the correct way when I do a PCA+LDA plot? Releated links: https://stackoverflow.com/questions/62573178/pca-lda-analysis-r Proportion of explained variance in PCA and LDA)

$\endgroup$
2
  • 1
    $\begingroup$ Do you mean that you are plotting both PCA and LDA axes in the same figure or that you are performing LDA on PCA scores? $\endgroup$
    – user289381
    Jun 27 '20 at 22:17
  • $\begingroup$ I wish to know what will be the correct way when I perform LDA on PCA data $\endgroup$ Jun 28 '20 at 10:27
0
$\begingroup$

If you are plotting the LDA points calculated from PCA scores, you just use the percentage of LDA explained variance.

In the figure caption or in the methods you describe the choice of the principal components and the details about the explained variance. You can see PCA as a feature engineering process, then LDA as a classification model based on them.

$\endgroup$
4
  • $\begingroup$ LDA present the % of explained between-group Variance but in this case I do not have groups just PC1 and PC2 (for example) so what the LDA % explain? $\endgroup$ Jun 28 '20 at 15:04
  • $\begingroup$ How would you apply LDA if you don’t have groups? $\endgroup$
    – user289381
    Jun 28 '20 at 15:11
  • $\begingroup$ script can be seen in stackoverflow.com/questions/62573178/pca-lda-analysis-r $\endgroup$ Jun 28 '20 at 15:54
  • $\begingroup$ Yes but if you look at the formula for lda, they use species as a grouping factor. If you have a grouping factor then you just use the pca scores as predictors, as explained in my answer. $\endgroup$
    – user289381
    Jun 28 '20 at 16:05

This site is temporarily in read only mode and not accepting new answers.

Not the answer you're looking for? Browse other questions tagged .