phenolology 🐢 @scisus@scholar.social

I should have run these models in the winter.

Sometimes making your stan model more complex makes it go WAY FASTER. https://mc-stan.org/docs/2_19/stan-users-guide/multivariate-hierarchical-priors-section.html

Stan really doesn't like my model constraints :( :( :( :( :(

Today I am grateful for the mute button

It's cool that cemeteries have a lot of rare plants, but we don't have to believe in Jesus to save them.

https://www.sciencedirect.com/science/article/pii/S235198941930174X

And now it's time for me to stop noodling at papers and start working on my own model :D

I'm trying to figure out if I need to have varying slopes AND intercepts in my ordered logistic model, so I'm going back to looking at the data and maybe building some simple models of transitions between pairs of categories.

What have you been up to lately?

Oh! This is a nice paper summarizing a lot of information about prior choice in Bayesian models. It's targeted at ecologists and the kinds of models they tend to use, but useful outside of ecology I expect.

https://onlinelibrary.wiley.com/doi/abs/10.1111/oik.05985

#ecology #modeling #stan #bayesian

This is an ambitious as heck paper and I am looking forward to reading it as soon as I can get access.

Understanding ecological change across large spatial, temporal, and taxonomic scales: integrating data and methods in light of theory

https://onlinelibrary.wiley.com/doi/abs/10.1111/ecog.04616

#ecology #macroecology

I love when science gets this specific.

So the air has a bunch of tiny particular matter in it - dust from the Sahara, soot from your car, etc. When it rains, some of it washes out of the air and into the forest. Does different PM and amounts of it wash out in different kinds of forests? Along the tree trunks vs in between the trees? Yes!

And this has implications for nutrient availability for the trees, among other things.

https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2019JD030276

So a colleague I met with last week had a really good and simple idea to help me figure out whether I need to complexify or decomplexify my model. Gonna try it today and see what happens!

If the cost of an industry outweighs its benefits, we should get rid of it. But how do we quantify those costs and benefits?

"This article provides a path to an objective quantifiable metric for determining when an entire industry warrants the corporate death penalty."

https://www.mdpi.com/2076-0760/8/2/62

I had a meeting earlier this week with a prof in my department about some modelling problems I'm having and it was really helpful and encouraging.

I finally got #stan to do a complicated thing I've been working on for awhile.

But not the cutpoints difference problem, unfortunately.

Anyway, a good step and I feel good about it.

Ok, today I edited my question on the stan forums to be clearer (and fixed an error in the code 😅 ). I've got an errand to do this morning and then I'm meeting with someone who might have some insight into my problem.

What are you up to today?

Hey if you're a #stan user or build ordered logistic models, halp me?

https://discourse.mc-stan.org/t/add-effects-to-difference-between-cutpoints-in-multilevel-ordered-logistic-model/8421

#statistics #modeling

I know the error is non-fatal,but can't you tell me more about the Jacobian treatment doc?

If it don't get this model right I could die anyway!

We really put our street trees through some tough situations!

https://www.sciencedirect.com/science/article/pii/S1618866718305910

The adaptive potential of plant populations in response to extreme climate events

https://onlinelibrary.wiley.com/doi/abs/10.1111/ele.13244?af=R

#adaptation #climatechange #plants