Choose Language Hide Translation Bar
Community Trekker

Best way to visualize non-normal (negative binomial) GLM?


I've got a (zero-inflated) negative binomial GLM that I'm using and am curious about visualizing this in JMP. Is it possible? I've got a few different models (all with negative binomial distributions); some have just the continuous predictor and the response variable while a few others have the predictor, a blocking factor, and the interaction. Overall, pretty simple models. How can I graph this in JMP? I've graphed it linearly to get an idea of the trends but would like to graph it with the proper distributions now.

Any input would be greatly appreciated! Thank you.

0 Kudos

Re: Best way to visualize non-normal (negative binomial) GLM?

Have you tried using the prediction profiler?



Each variable the model formula references needs to be created as a column.

You'll need one row for the high value, and one row for the low value for each of the input columns.

Finally, paste your model into a column as a formula. Make sure the variable names match the column names.



JMP Systems Engineer, Pharm and BioPharm Sciences
0 Kudos