search

squidgroup / squidSim_vignette

vignette for the updated squid R package
0 stars 0 forks source link

last example in http://squidgroup.org/squidSim_vignette/2.4-randomslopes.html #7

Open BertvanderVeen opened 2 years ago

BertvanderVeen commented 2 years ago

http://squidgroup.org/squidSim_vignette/2.4-randomslopes.html includes an "interactions" argument in the simulations, yet there is no interaction included in the model. This term seems to be used to set-up random-slope model, which includes a multiplication of a parameter vector, and a predictor, which is not an "interaction" as in the classical fixed-effects sense. Thus, the "interaction" terminology might be confusing.

joelpick commented 2 years ago

The distinction here is more a modelling one rather than a simulation one - in the simulation it is an interaction (i.e. two variables multiplied), it is just with an unobserved variable (as commented on in the first line of that vignette section). I'll flesh out the text so it is more clear.

This is also described as an interaction in other place (for example see chapter 4 here https://mran.microsoft.com/snapshot/2018-08-24/web/packages/MCMCglmm/vignettes/CourseNotes.pdf)

BertvanderVeen commented 2 years ago

Yep, I understand that. However, most people will approach this from a modelling perspective (few ecologists are familiar with simulation, right?). So my point was that it might be good to document what "interaction" in this context means.

What exactly am I looking for in those course notes?

It is news to me that the term "interaction" is used outside of modeling, wikipedia seems to agree with me: https://en.wikipedia.org/wiki/Interaction_(statistics). An interaction is not synonymous to the product of any two (or more) terms in mathematics/statistics, hence the confusion for me (and likely others) here. Could something along the lines of "products" (of terms) or so be a more suitable alternative?