- `cbind(NULL)` returns `NULL` (not a matrix). For consistency
`rbind(NULL)` also returns `NULL`.
Explore
103,902 skills indexed with the new KISS metadata standard.
`rbind(NULL)` also returns `NULL`.
not position. Missing columns get `NA`.
class table. Convert to data frame with `as.data.frame(tbl)`.
not `cbind.matrix`. Mixing matrices and data frames can give unexpected results.
~ Month)`.
interaction is used; levels containing `.` can cause unexpected splits unless `sep` is changed.
so FUN should not expect additional args with same length as X.
adding a Row.names column.
names(y))` — can silently merge on unintended columns if data frames share column names.
Date objects become numeric).
`MARGIN`
matrix
result has dim `c(n
`apply` coerces to matrix via `as.matrix` first — mixed types become character.
raw vectors
j)]`) coerces to matrix first — avoid.
1]` returns a **vector** (drop=TRUE default for columns)
which can silently pick up wrong variables in programmatic use. Use `[` with explicit logic in functions.
i = 1]` is `m[2
gotchas
no default. `s.window = periodic` assumes fixed seasonality.
`trend`
uses moving averages. `type = additive` or `multiplicative`.
package = pkg)` opens a specific one.