theoreticalecology / advancedregressionmodels Goto Github PK

Add glmmTMB example to pgls structure

Abschnitt über Hypothesis translation in model selection chapter

-> erklären wie man eine wissenschaftliche Hypothese in eine Formel umsetz

• Formel
• Centering / Scaling
• Contrasts
• ...

Ist das beabsichtigt? (Findet man in fast allen Chapters)

See also https://jamanetwork.com/journals/jamapsychiatry/article-abstract/481967

https://economics.stackexchange.com/questions/24010/difference-between-multinomial-logit-and-conditional-logit#:~:text=Typical%20explanations%20of%20how%20the,function%20of%20the%20choices'%20characteristics.

To add

https://www.cell.com/current-biology/fulltext/S0960-9822(16)30670-4#%20

https://x.com/jedmsp/status/1720522455020662975?s=46

Basierend auf airquality

a) summary + str
b) chr oder numeric to faktor umwandeln (airquality)
c) skalieren, transformieren, neue Variable im DF erstellen
d) subsetted, NAs remove, complete.cases
e) plotten (numeric gegen numeric, numeric gegen factor, hist), pairs
f) correlation?

Add comments on

• error in variable models
• latent variables
• integrated models

https://solomonkurz.netlify.app/blog/

Expand on GLMM estimation

• REML motivation and relationship to Bayes, see https://nebula.wsimg.com/97996deee1d889c3063478a26ebedf1c?AccessKeyId=8092329B7EC3FCEF893E&disposition=0&alloworigin=1
• https://cran.r-project.org/web/packages/lme4/vignettes/Theory.pdf

https://bookdown.org/mike/data_analysis/

https://declaredesign.org/r/fabricatr/articles/other_packages.html

https://twitter.com/m_wegmann/status/1580420625268473859

https://www.sciencedirect.com/science/article/pii/S0749596X17300013, pro selection on RE structure

Add warning, see also arguments here https://academic.oup.com/book/33545/chapter/287918750

Following

https://lindeloev.github.io/tests-as-linear/

Grafik mit

0 0 0
0 1 0
0 0 1
...

für die verschiedenen Contrasts um zu erklären was passiert

https://onlinelibrary.wiley.com/doi/full/10.1002/ece3.9062 section 2.4

https://stats.stackexchange.com/questions/115090/why-do-i-get-zero-variance-of-a-random-effect-in-my-mixed-model-despite-some-va?noredirect=1&lq=1

https://stats.stackexchange.com/questions/323273/what-to-do-with-random-effects-correlation-that-equals-1-or-1