Ben Bolker

bbolker.bsky.social

21 July 2026 15:47

For the record I said that the (1|g) term is redundant in the *second* term, i.e. (1 | g) + cs(0 + f | g, hom = TRUE) ... (1|g/f) is identical to (1|g) + (1|g:f) (or (1|g) + (1|f:g)), and the (1|g) is *not* redundant ...

Emmanuele Tidoni

letstido.bsky.social

21 July 2026 15:27

(2) if I am not wrong, mdl1 has better AIC/BIC/logLik than mdl4, but a similar comparison does not differ here (see table1 where AIC difference is 1.99): cran.r-project.org/web/packages... so, probably the relevance (or not) of 1|g depends on the dataset(?)

Emmanuele Tidoni

letstido.bsky.social

21 July 2026 15:27

(1) When we started those series of studies, that was our initial thought too (that some models would have resulted as redundant). Simulations showed that 1|g is not redundant as models without 1|g had low power (table 2 of the paper).

Luis Endera

luisendera.bsky.social

20 July 2026 10:29

Su madre, María Arthuer, cruzó el desierto del Sahara a pie y descalza en 1994. Saltó la valla de Melilla embarazada de Iñaki y sin recursos. Hoy su hijo, es campeón del mundo. #respect

Nicola Miller

nicmiller.bsky.social

20 July 2026 10:28

These photos fill me with joy.