Handling multiple observation types in a single series / time about mvgam HOT 2 CLOSED

Hi @matt-w-rees, that is an interesting problem. I'd be most tempted to treat each trap type as a different 'series' and then perhaps let them share the same latent process model (you can see how to do this in the shared states vignette: https://nicholasjclark.github.io/mvgam/articles/shared_states.html. You could then include an effect of trap type in the observation formula, which would hopefully allow you to readily identify the latent process. But you might have to give it a go and let me know what problems you run into, if any!

from mvgam.

Thanks Nicholas - I hadn't thought of that approach. I'm not sure it is worthwhile to add another series given there isn't too many times this occurred, so might just add this to the list of things to try down the track.

from mvgam.