General
PromptBeginner5 minmarkdownQuality: 28
- For goodness-of-fit: pass a vector
not a matrix. p must sum to 1 (or set rescale.p = TRUE).
0
Explore
Find agent skills by outcome
138,431 skills indexed with the new KISS metadata standard.
Showing 24 of 138,431Categories: Data & Insights, Coding & Debugging, Data, Creative, General, Openclaw
General
PromptBeginner5 minmarkdownQuality: 28
- uniroot: finds a root of f in `[lower
upper]. Requires f(lower) and f(upper)` to have opposite signs.
0
General
PromptBeginner5 minmarkdownQuality: 24
Non-obvious behaviors
gotchas
0
General
PromptBeginner5 minmarkdownQuality: 24
- `anova(model1
model2)`: F-test comparing nested models.
0
Coding & Debugging
PromptBeginner5 minmarkdownQuality: 28
- SANN (simulated annealing): convergence code is always 0 — it never fails. maxit = total function...
no other stopping criterion.
0
General
PromptBeginner5 minmarkdownQuality: 24
- Returns object with predict
0
General
PromptBeginner5 minmarkdownQuality: 28
- Default method is Nelder-Mead (no gradients
robust but slow). Poor for 1D — use Brent or optimize().
0
General
PromptBeginner5 minmarkdownQuality: 28
- predict.glm with type = response gives predictions on the response scale (e.g.
probabilities for logistic); type = link (default) gives on the link scale.
0
General
PromptBeginner5 minmarkdownQuality: 24
- lowess parameter is f (span
default 2/3); loess parameter is span (default 0.75).
0
General
PromptBeginner5 minmarkdownQuality: 24
- lowess is the older function; returns `list(x
y)` — cannot predict at new points.
0
General
PromptBeginner5 minmarkdownQuality: 24
- Direction: direction = both (default)
forward
0
Data
PromptBeginner5 minmarkdownQuality: 28
- If data fits too exactly (no residual noise)
convergence check fails — use control = list(scaleOffset = 1) or jitter data.
0
Creative
PromptBeginner5 minmarkdownQuality: 24
- Self-starting models (SSlogis
SSasymp
0
Creative
PromptBeginner5 minmarkdownQuality: 28
- aov is a wrapper around lm that stores extra info for balanced ANOVA. For unbalanced designs
Type I SS (sequential) are computed — order of terms matters.
0
General
PromptBeginner5 minmarkdownQuality: 24
- For Type III SS
use car::Anova() or set contrasts to contr.sum/contr.helmert.
0
General
PromptBeginner5 minmarkdownQuality: 24
- Quasi-families (quasibinomial
quasipoisson) allow overdispersion — no AIC is computed.
0
General
PromptBeginner5 minmarkdownQuality: 24
- binomial accepts response as: 0/1 vector
logical
0
General
PromptBeginner5 minmarkdownQuality: 28
- `anova(glm_obj
test = Chisq) for deviance-based tests; F` is invalid for non-Gaussian families.
0
General
PromptBeginner5 minmarkdownQuality: 28
- weights in glm means prior weights (not frequency weights) — for frequency weights
use the cbind trick or offset.
0
General
PromptBeginner5 minmarkdownQuality: 24
- Column names from model.matrix can be surprising: e.g.
factorLevelName concatenation.
0
General
PromptBeginner5 minmarkdownQuality: 24
- Common families: binomial(link = logit)
poisson(link = log)
0
General
PromptBeginner5 minmarkdownQuality: 28
- Default family = gaussian(link = identity) — glm() with no family silently fits OLS (same as lm
but slower and with deviance-based output).
0
General
PromptBeginner5 minmarkdownQuality: 28
- I() is required to use arithmetic operators literally: y ~ x + I(x^2). Without I()
^ means interaction crossing.
0
General
PromptBeginner5 minmarkdownQuality: 24
- terms object attributes: order (interaction order per term)
intercept
0