[R-sig-eco] ordiR2step error

[Marika Galanidi]

Dear all,

after reading the Legendre & Anderson (1999) paper again I realised I had
not included a correction for negative eigenvalues in my models. Square root
transformation of the distances did not eliminate negative eigenvalues but
adding a constant did the trick. Apologies for rushing to resort to the
forum, I should have done my homework better...

Regards,
Marika



--
View this message in context: http://r-sig-ecology.471788.n2.nabble.com/ordiR2step-error-tp7579341p7579342.html
Sent from the r-sig-ecology mailing list archive at Nabble.com.