search

lizzieinvancouver / ospree

Budbreak review paper database
3 stars 0 forks source link

Write up methods for ranges paper #420

Open cchambe12 opened 2 years ago

cchambe12 commented 2 years ago

@cchambe12 @dbuona @MoralesCastilla We should all work on writing up methods to prepare for the ms and hopefully include captions for relevant figures.

lizzieinvancouver commented 2 years ago

@cchambe12 @dbuona Is there anything I should write up? I am thinking no.... but realized that I should check.

cchambe12 commented 2 years ago

@dbuona I just wrote up the methods for my section. I'm not totally sure if I wrote the model equation correctly. I'd be happy to work with someone if folks think it is incorrect but I could certainly use some help!

lizzieinvancouver commented 2 years ago

@cchambe12 Happy New Year!

I was doing a quick look at your equations in the range methods and i think they're a little off, but I wanted to check.

You wrote:

y & \thicksim N(\alpha_{study[i]} + \alpha_{sp[i]} + \alpha_{pop[i]} +& \beta_{forcing_{sp[i]/pop[i]}} + \beta_{photoperiod_{sp[i]/pop[i]}} + \epsilon_[i]} \\

Which implies intercept for study and a SEPARATE intercept for species, and a SEPARATE intercept for population, but I thought population was nested in species? I also suspect we may have had a grand intercept ... so something more like y \thicksim \alpha_0 + \alpha_{study[i]} + \alpha_{sp[pop[i]]} ... is what I think we might have?

Also, sp[i]/pop[i] is lmer notation, so not mathematically clear. Here's a simple example of how that should look for an intercept only model:

y_{i}=\alpha_{sp[pop[i]]}}+\epsilon_{i}

\alpha_{sp[pop]} \sim N(\mu_{sp}, \sigma_{sp})

\mu_{sp} \sim N(\mu_{pop}, \sigma_{pop}})

But if you have a grand intercept, then you would have a 0 in there somewhere. Can you take another stab at it and then point me to the Stan code and I can check again (though I am terrible at notation myself, so we'll need another set of eyes in the end).

lizzieinvancouver commented 2 years ago

@dbuona @cchambe12 I updated the methods for the OSPREE database and for how we picked species ... but I do not think I am sure how we picked species. I think we have to check both our code and notes together to be sure. So, can someone check both the code and the relevant issue and update the methods text? And then maybe the other person could double-check?

Thanks!

cchambe12 commented 2 years ago

@lizzieinvancouver Okay, I have gone through the methods for the sppop model equation and have updated according to your notes - THANK YOU! Makes a lot of sense and I think it should be much closer now.

@lizzieinvancouver @dbuona I have updated the section on species selection. Dan, would you mind reading through it and making sure it looks okay? I have also added information on the species maps and citations for these.

@MoralesCastilla I have added in info on the climate data for North America - thanks for writing this section up!