Im getting significant deviations in KS tests for a glmm with guassian distribution

florianhartig / DHARMa

Diagnostics for HierArchical Regession Models

http://florianhartig.github.io/DHARMa/

200 stars 21 forks source link

Im getting significant deviations in KS tests for a glmm with guassian distribution #397

Closed ReneeAWB closed 6 months ago

ReneeAWB commented 6 months ago

Hello! Im new to using R and im having trouble with a few things. I am working on finding out whether age and sex affects the weight of sheep. I am using the guassian disrtibution and when i use DHARMa, the KS test shows this warning. Im not sure if this is a significant issue or not, and im having trouble trying to find ways to fix it.

florianhartig commented 6 months ago

Hi,

please see remarks in https://cran.r-project.org/web/packages/DHARMa/vignettes/DHARMa.html#interpreting-residuals-and-recognizing-misspecification-problems , in particular point 3.

In your case, the KS test is significant, but there is obviously although significant NO substantial deviation from the assumptions in your plot (because the QQ line is practically straight). So, I see no concern in your plot.

Best F