ercbk.bsky.social
@ercbk.bsky.social
Github: https://github.com/ercbk
Website: https://ericbook.netlify.app/
Mastodon: @erc_bk@fosstodon.org
Whether it was you or Claude, nice work figuring out how to aggregate to those H3 cells. Thought maybe I could shorten the code in that function, but nope. Also, would've thought that aggregating would be something in the docs, but I couldn't find anything while skimming {h3o} or {h3jsr} docs.
November 26, 2025 at 11:40 PM
Ahh I see. Thanks.
November 21, 2025 at 1:58 PM
If I'm understanding correctly, the Ohio State value is nonsensical given they're undefeated.
November 21, 2025 at 1:30 PM
I hate reading code that uses it. It makes it more difficult to find assignments. Instead of just skimming the left side of the script, I also have to look at the right side and find the end of piped code.
November 19, 2025 at 1:38 PM
Friendly's book is also online. (It's posted at his uni, so I assume it's legal.) See Ch.11.
euclid.psych.yorku.ca/www/psy6136/...
November 12, 2025 at 2:42 PM
Are there any run/passpro PFF grades for Michalski this season?
November 1, 2025 at 7:25 PM
I like how the background of the value label matches associated sunburst(?) segment. Didn't see anything special in the code to do that. Does ggrepel do that by default?
October 23, 2025 at 1:04 PM
Harrell has a biostatistics for biomedical researchers video course
Homepage - hbiostat.org/bbrc/
Youtube channel - www.youtube.com/@bbrcourse62...
Notebook - hbiostat.org/bbr/
October 7, 2025 at 11:11 PM
Schoch has a few. github.com/schochastics...

This typst one is also pretty nice. Quarto allows you to use typst templates, so I'd think it'd be possible to use this.
github.com/peterpf/mode...
quarto.org/docs/referen...
October 4, 2025 at 5:39 PM
Going from PO to PPO is not that big of a leap in terms of fitting and interpretation from my understanding, so maybe those would be helpful actually. Big fan of yuzaR.
September 28, 2025 at 12:06 PM
Those are really nice proportional odds videos, but nothing to do with partial proportional odds.
September 28, 2025 at 11:57 AM
Also "'cat I farted' machine" is hilarious. Is that Aussie for chatgpt?
September 26, 2025 at 1:57 PM
He has a datamethods forum where it looks like its been mentioned quite a few times, discourse.datamethods.org/search?q=par.... He's also on bluesky and pretty responsive to questions.
Search results for 'partial proportional odds' - Datamethods Discussion Forum
This is a place for discussions and Q&A about data-related issues and quantitative methods including study design, data analysis, and interpretation.
discourse.datamethods.org
September 26, 2025 at 1:54 PM
It is Frank Harrell's model, and he has one article that (I know of) he refers to about it (besides the original paper), hbiostat.org/blog/post/yb.... I only speak a little Harrell-ese and don't use his pkgs much, but maybe it can help you.
Borrowing Information Across Outcomes – Statistical Thinking
In randomized clinical trials, power can be greatly increased and sample size reduced by using an ordinal outcome instead of a binary one. The proportional odds model is the most popular model for ana...
hbiostat.org
September 26, 2025 at 1:47 PM
Yeah, I don't have many resources on the partial proportional odds model either. What I know about it is mostly from the {ordinal} vignette. But it does relax the PO assumption, so you're in the right area.
September 26, 2025 at 1:37 PM
This is what I get from gtsummary::tbl_regression(nom_mod, intercept = TRUE, exponentiate = TRUE). Are you not setting intercept = TRUE?
September 26, 2025 at 1:28 PM
*shown in gtsummary::tbl_regression
September 25, 2025 at 1:14 PM
Not clear what you mean by "only for the predictors that aren't specified as nominal." This is a PPO model, so location doesn't get a regression coef. The partial effects (thresholds|locaton) are exp and shown. So besides CIs for partial effects, I'm not sure what you're looking for.
September 25, 2025 at 1:12 PM
Good Kamara take
September 17, 2025 at 12:53 PM
Didn't get to*
September 17, 2025 at 12:38 PM
Did get to explore the report since I don't have any soil data, but I liked the design of the landing and learn more pages. The side panel with clickable step 1,..., step4 was cool. Is that custom?
September 17, 2025 at 12:36 PM
The kickoffs stopped going in the endzone. Did IU switch to the back-up after halftime?
September 13, 2025 at 1:19 AM
Try using double backticks.
September 5, 2025 at 4:40 PM