Project Covariates onto DISTATIS Maps

Usage

# S3 method for class 'distatis'
project_covariate(
  x,
  y,
  side = c("table", "feature"),
  what = "dimension",
  scale = c("cosine", "beta"),
  ...
)

Arguments

x: A fitted distatis model.
y: Finite numeric covariate, one value per assessor or object.
side: "table" for assessors or "feature" for objects.
what: "dimension" returns uncentered columnwise cosines or regression slopes in the selected map. For table covariates, "feature" (alias "observation") returns an object-by-component pattern by projecting the partial scores across assessors onto y.
scale: "cosine" or "beta". For dimension projections the slope regresses y on each map axis. For feature patterns it regresses each partial-score coordinate on y. No intercept or automatic centering is used.
...: Unused arguments, rejected.

Value

A named vector for dimension projections, otherwise an object score matrix. Assessor and object projections may have different axis counts.