Mgcv has a hard time if your independent smooth variables are anything but randomly distributed. Honestly, I would just use a random forest. Then all the interactions are naturally incorporated and you aren't committing to linear relationships with lat and long.
21.11.2023 20:26 β π 0 π 0 π¬ 1 π 0
How are you fitting your GAM? Maybe some regularization is needed? Are you wedded to GAM? A Random Forest would be self regularizing within the sample outcome ranges.
21.11.2023 13:34 β π 1 π 0 π¬ 1 π 0
Kind of a shot in the dark, but does anyone have any connections who do homeowner's insurance that I could speak with on the phone? I don't want to buy, I just want to ask some questions about the process to someone who doesn't have motives.
12.11.2023 00:41 β π 0 π 0 π¬ 0 π 0
Text analysis using AI in R
Introduction Analyzing qualitative data is challenging. Such analyses are even more difficult when the topic is controversial and the results will drive important policy decisions. This post explorβ¦
Happy to release this blog post demonstrating an end-to-end workflow for qualitative text analysis using ChatGPT classifications, openAI embeddings to uncover hidden topic via clustering, and context summarization with Claude2. All in #rstats.
www.azandisresearch.com/2023/10/05/t...
18.10.2023 11:53 β π 4 π 2 π¬ 0 π 0
If you can't afford the application, you definitely can't afford grad school salaries.
25.09.2023 10:35 β π 1 π 0 π¬ 0 π 0
Tell me who to follow to make this not true.
19.09.2023 11:47 β π 2 π 0 π¬ 0 π 0
Made the original For You feed
π§π· | Baiano perdido no DF | Software Engineer | All photos taken by me π· | renoviana.com
π¦ @avesbrasil.bsky.social
π @livro.bsky.social
Writer (The Atlantic). Current π: SUMMER OF OUR DISCONTENT (Knopf). Non-resident fellow (AEI). Senior fellow, visiting prof. of humanities (Hannah Arendt Center, Bard College). Representation (APB; Wylie). Student & Teacher. Instagram: @chattertonwilliams.
Exploring the American idea through ambitious, essential reporting and storytelling. Of no party or clique since 1857. http://theatlantic.com
social psychology βͺοΈ metascience βͺοΈ philosophy of science βͺοΈ higher education
Professor at Durham University, UK. He/him.
Website: https://sites.google.com/site/markrubinsocialpsychresearch/
Substack: https://markrubin.substack.com/
Metascience, statistics, psychology, philosophy of science. Eindhoven University of Technology, The Netherlands. Omnia probate. πͺπΊ
Psychology prof at the University of Toronto.
Genetics. Evolution. History
https://www.razibkhan.com
https://www.razib.com
evolutionary bio postdoc at HΓ³lar University.
fish stuff mostly, but not exclusively.
Skip emeritus of Colgate University curling team.
Formerly: McGill, W&M, Colgate
Bethlehemite (PA)
gehaines.weebly.com
Bad Astronomer (formerly @BadAstronomer), SciCommer, Dork. Crash Course Astronomy guy. He/him. Newsletter: badastronomy.beehiiv.com
I'm self-employed so my opinions do in fact represent my company's.
Solitary, poor, nasty, brutish and short.
PhD student at UConn | Analyst at the Medical Evidence Project | Center for Scientific Integrity| matthewbjane.com
Betz Chair of Ecology at Drexel University in Philadelphia. Urban ecology & evolution, landscape genetics, native plants, rats, lichens. Problematic political opinions all my own!
Building data science tools @posit.co, #rstats, parenthood β¨πβ¨
Lover of event history analysis, causal inference, Bayesian inference, and their intersection. Also a lover of representative democracy. Posts are my own thoughts. This is a personal account. In medio tuttisimus ibis.
Visualisation and graphics @posit.co
Classic Generative Art Weirdo using π€ and R: http://thomaslinpedersen.art and http://deca.art/thomasp85
he/him
Data Science, Causal Inference, Economics, Statistics and Machine Learning
Two joys in life:
[1] Learning something interesting about the world
[2] Telling people about it
Data Science @OpenAI
Building. Affiliations: @JHU, @Penn, @UCSC, @Amazon, @Twitter || Art: #NLProc, Vision, Speech, #DeepLearning || Life: ιε
, improv, running π
