Skip to content
Prev 9760 / 20628 Next

Error en mer_finalize(ans) : Downdated X'X is not positive definite, 1. What is wrong with my model?

Dear Ben,

Thanks for your response. I had read a very similar response to FAQs  
explaining why the error relates to rank deficient models. I have  
tried to center Treatment (a continuous variable, sorry I didn't  
specify this!) but this does not remove the error.

I think the problem comes with the nested nature of "Species" within  
"Bud_type", since not all species are in all Bud_types, I have a rank  
deficient design and lme4 cannot cope with it. I  do not think there  
is a solution to this problem other than including "Species" as a  
random effect, but then I will not be able to know its effect...

Cheers,

Sara Palacio

Quoting Ben Bolker <bbolker at gmail.com>: