Collin Edwards

@[email protected]
192 Followers
210 Following
306 Posts
PhD in ecology and evolutionary biology. Mathematical and statistical ecologist, R programmer. Works at the interface of data and theory to explore population dynamics, species interactions, community assembly. Blacksmithing and martial arts enthusiast. He/Him.
Websiteevo-eco.org/
githubgithub.com/cbedwards
Google Scholarhttps://scholar.google.com/citations?user=SI50LP8AAAAJ
Show threadCollin EdwardsMar 29, 2024
@Lluis_Revilla Yep, this is the solution I've been working towards. I've been using openxlsx::loadWorkbook to maintain formatting wrt merged cells, cell borders, bold text for subheaders. I'm running into some performance issues, though, as I need to set all cells to have white background before I highlight cells that changed, and it looks like addStyle doesn't scale especially well with large numbers of cells. But I'll experiment a bit.
Show threadCollin EdwardsMar 24, 2024
@frod_san This might be exactly what I want! Thanks!
Show threadCollin EdwardsMar 24, 2024
@Lluis_Revilla These are definitely options, but some of the files I want to compare have dozens of worksheets in them, which could get tedious. And the ideal use-case is a tool that less coding-savvy teammates can use, so outputting to excel with highlights on cells that changed would be a big benefit
Collin EdwardsMar 22, 2024
In my new position I'm frequently needing to compare different versions of the same excel file (the FRAM model uses excel for inputs and outputs; I frequently need to compare the inputs or outputs of different model runs). Does anyone know of a good non-browser tool for comparing excel files? Basically looking for a `diff` analog. I'm tempted to write an #Rstats package for this, but if it exists I'd rather not re-invent the wheel.
Collin EdwardsJan 29, 2024
R-Ladies RomeJan 29, 2024

Next programmed event for @rladiesrome:

Title: Debugging in R
Speaker: Shannon Pileggi - @PipingHotData
When: Feb 20th 6.15pm CET
RSVP: https://www.meetup.com/rladies-rome/events/298710129/
#rladies #rstats #debugging #datascience

@fgazzelloni @RConsortium @RLadiesGlobal

Debugging in R, Tue, Feb 20, 2024, 6:15 PM | Meetup

Join [R-Ladies Rome](https://rladiesrome.quarto.pub/website/) for an empowering **Interactive Workshop**: **Debugging in R** exclusively designed for women in technology. I

Meetup
Collin EdwardsJan 29, 2024
Peter GleickJan 25, 2024

Tip of the Day!

Here is a reminder to everyone [rightly] pissed at the incredibly high cost of peer-reviewed journal articles behind paywalls:

If you send an email to the lead/corresponding author (whose contact is almost always provided in the article, or easily found with google) and ask for a copy of the article, the author will almost universally be delighted to send a copy [legally] to you, free.

We authors LOVE it when someone asks for a copy of our articles!!!

Collin EdwardsDec 9, 2023

I'm excited to share a preprint of my work identifying statistical methods for modeling butterfly populations in #ecology and #conservation. I developed guidelines for fitting sparse transect monitoring data with Generalized Additive Models (GAMs), based in large part on a new #Rstats package I developed to simulate and fit realistic butterfly data.

https://biorxiv.org/cgi/content/short/2023.12.07.570644v1

Show threadCollin EdwardsJul 28, 2023
@ChristosArgyrop @noamross @gavinsimpson EVEN REVIEWER 2??
Show threadCollin EdwardsJul 28, 2023
@ChristosArgyrop @noamross @gavinsimpson Hahaha, definitely we're already using AIC. My thinking is to do double duty by also providing P values (with caveats), just so that we don't have less statistically aware reviewers turn down the analysis.
Show threadCollin EdwardsJul 28, 2023

@gavinsimpson @noamross This is awesome!! Thanks for all the help!

We're already doing comparisons with AIC, and the differences are quite large. I was just thinking providing a reasonable P value might help with the review process. So I'm not super worried that the P values are biased towards being low.