Multispecies / multiresponse vignette

[ericward-noaa]

I talked to Lewis last week about this. Currently includes examples of:

• main effects by species
• spatial fields by species, with and without shared variance
• spatiotemporal fields by species

Would be good to get any other examples added @Lewis-Barnett-NOAA @seananderson

[Lewis-Barnett-NOAA]

Looks great! I just did a little editing for clarity, flow, and style in a commit I just pushed. I also added a few lines to reveal the results of the model fits, sdreports, and sanity checks.

Is there a better way to display the output of sanity() in this format? I'm not getting the pretty check marks and all that good stuff. Otherwise I think we are good to go. It would be better of course if we had some models that actually fit quite well, but at least this shows the progression of performance among approaches.

[seananderson]

Thanks for this! I did a bunch of editing. Among other things, the models all converge now (a main reason was one of the species was simulated with Tweedie error by accident and I increased the sigma_E to keep it from collapsing). I added some more explanation around the mapping. I added a section on hacking species or other groups into time for spatial models. I removed the cohort part at the end for now, but alluded to how the same approach can be used for composition standardization. I started an example over here and I'd like to ultimately make a vignette on that. Hopefully soon.