I am using DFA to analyze my data in R.

In SPSS it gives Wilks' $\lambda$ with the output, but in R I couldn't determine how to compute it. I have found code that is available to calculate the Wilks' $\lambda$ value for MANOVA.

I am new to R and it would be helpful if anyone can help me to compute $\lambda$ value for the LDA in R.

  • 4
    $\begingroup$ The man's name is Wilks, not Wilk. Wilks' lambda is the proportion of within-groups sum of squares to total sum of squares. Take these SS from ANOVA table for discriminant scores. $\endgroup$ – ttnphns Nov 18 '12 at 12:28

I found good information on the Wikipedia page on Linear Discriminant Analysis but the English page is not as clear as the French one where they give the formula for both Wilks' lambda and Fisher's F.

Wilks' lambda is the ratio of the determinants from the within-group covariance matrix to the total covariance matrix $\cfrac{|W|}{|V|}$, $W$ (resp. $V$) being the restriction of the within-group (resp. global) covariance matrix to the currently selected parameters.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.