CODEWITHSUNDEEP
rglm
Gabriel Johnson
Gabriel Johnson

Reputation: 1

I need help to debug my Zeroinfl GLM that causes an error indicating 'NA' and "exactly singular systems"

I am having a problem with my code. I am trying to create a GLM model with Zeroinfl since my data set has many 0's in it. I am trying to study the influence of pond characteristics on pond amphibian diversity but I get an error code that leads to "NA"'s in my summary table.

When I try to create my model :

modele_amphibia <- zeroinfl(diversite_sp_10 ~ surface + date + 
                   substrat + res_hydro + vol_pot_est+
                   creation_restauration + habitat, 
                   data = data_GLM_amphibia, dist = "poisson")

I get this error :

Message d'avis : Dans value[3L] : Routine Lapack dgesv : le système est exactement singulier : U[22,22] = 0FALSE

I don't understand how I can fix this.

When I do summary(modele_amphibia)

I get my coefficient numbers but "NA" in all the other columns...

What can I do ?

Upvotes: 0

Views: 22

Answers (0)

Related Questions