I was hoping I could ask you a few questions about using DHARMa for Bernoulli Response Data. My first question has to do with the calculation of the residual. Since the observed values are either 0's or 1's, how does the comparison of the observed value with the ECDF generate the residual? I am sure I am getting mixed up somewhere, but wouldn't this just generate residuals that are either 0 (if the observed value is zero) or 1 (if the observed value is 1)?
Secondly, can DHARMa be used to check the linearity assumption for Bernoulli response data?
Hi Florian,
I was hoping I could ask you a few questions about using DHARMa for Bernoulli Response Data. My first question has to do with the calculation of the residual. Since the observed values are either 0's or 1's, how does the comparison of the observed value with the ECDF generate the residual? I am sure I am getting mixed up somewhere, but wouldn't this just generate residuals that are either 0 (if the observed value is zero) or 1 (if the observed value is 1)?
Secondly, can DHARMa be used to check the linearity assumption for Bernoulli response data?
Thanks!