coef.svem_model: Coefficients for SVEM Models

Description

Extracts averaged coefficients from an svem_model fitted by SVEMnet.

Usage

# S3 method for svem_model
coef(object, debiased = FALSE, ...)

Value

A named numeric vector of coefficients (including the intercept).

Arguments

object: An object of class svem_model, typically returned by SVEMnet.
debiased: Logical; if TRUE and debiased coefficients are available (Gaussian fits with parms_debiased), return those instead of parms. Default is FALSE.
...: Unused; present for S3 method compatibility.

Details

For Gaussian fits, you can optionally request debiased coefficients (if they were computed and stored) via debiased = TRUE. In that case, the function returns object$parms_debiased. If debiased coefficients are not available, or if debiased = FALSE, the function returns object$parms, which are the ensemble-averaged coefficients across bootstrap members.

For Binomial models, debiased is ignored and the averaged coefficients in object$parms are returned.

This is a lightweight accessor around the stored components of an svem_model:

parms: ensemble-averaged coefficients over bootstrap members, on the model's link scale;
parms_debiased: optional debiased coefficients (Gaussian only), if requested at fit time.

Passing debiased = TRUE has no effect if parms_debiased is NULL.

Examples

Run this code

# \donttest{
  set.seed(1)
  n  <- 200
  x1 <- rnorm(n)
  x2 <- rnorm(n)
  eps <- rnorm(n, sd = 0.3)
  y_g <- 1 + 2*x1 - 0.5*x2 + eps
  dat_g <- data.frame(y_g, x1, x2)

  # Small nBoot to keep runtime light in examples
  fit_g <- SVEMnet(y_g ~ x1 + x2, data = dat_g, nBoot = 30, relaxed = TRUE)

  # Ensemble-averaged coefficients
  cc <- coef(fit_g)
  head(cc)

  # Debiased (only if available for Gaussian fits)
  ccd <- coef(fit_g, debiased = TRUE)
  head(ccd)

  # Binomial example (0/1 outcome)
  set.seed(2)
  n  <- 250
  x1 <- rnorm(n)
  x2 <- rnorm(n)
  eta <- -0.4 + 1.1*x1 - 0.7*x2
  p   <- 1/(1+exp(-eta))
  y_b <- rbinom(n, 1, p)
  dat_b <- data.frame(y_b, x1, x2)

  fit_b <- SVEMnet(y_b ~ x1 + x2, data = dat_b,
                   family = "binomial", nBoot = 30, relaxed = TRUE)

  # Averaged coefficients (binomial; debiased is ignored)
  coef(fit_b)
# }