Skip to contents

Expand a symbolized_model to its underlying numeric arrays

Source: R/expand.R
expand.Rd

expand() returns the actual numeric pieces that make up the fitted model: the response vector, design matrix per submodel, coefficient vectors, random-effect indicator matrix and BLUPs (when present), residuals, and the fitted conditional mean and SD vectors. This is the data side of the symbolized_model surface — the symbolic and matrix views show what shape the model has; expand() shows what numbers actually flow through it.

Returned object is classed c("symbolized_expanded", "list") so it has a compact print() method.

Usage

expand(x, ...)

Arguments

x

A symbolized_model.

...

Reserved for future use.

Value

A list with elements (NULL when not applicable to this fit):

  • y response vector of length n

  • X mu-submodel design matrix (n x p_mu)

  • beta mu coefficient vector (p_mu)

  • X_sigma sigma-submodel design matrix (n x p_sigma)

  • gamma sigma coefficient vector (p_sigma)

  • Z_g random-effect indicator matrix (n x G), if RE present

  • u random-effect BLUPs (G), if RE present

  • e residuals (n)

  • mu_hat fitted conditional mean (n)

  • sigma_hat fitted per-observation SD (n)