- `lowess` is the older function; returns `list(x
y)` — cannot predict at new points.
Explore
143,141 skills indexed with the new KISS metadata standard.
y)` — cannot predict at new points.
convergence check fails — use `control = list(scaleOffset = 1)` or jitter data.
`forward`
`SSasymp`
use `car::Anova()` or set contrasts to `contr.sum`/`contr.helmert`.
Type I SS (sequential) are computed — order of terms matters.
`quasipoisson`) allow overdispersion — no AIC is computed.
logical
test = Chisq)` for deviance-based tests; `F` is invalid for non-Gaussian families.
use the cbind trick or offset.
but slower and with deviance-based output).
`factorLevelName` concatenation.
`poisson(link = log)`
`^` means interaction crossing.
`intercept`
`contr.poly` for ordered.
`print`
gotchas
reserved = FALSE)` by default does NOT encode reserved chars (`/`
format = f
not factor.
integer
`left`
digits