Stepwise model selection for HubbellGLM models

Usage

# S3 method for class 'HubbellGLM'
step(
  object,
  scope,
  scale = 0,
  direction = c("both", "backward", "forward"),
  trace = 1,
  keep = NULL,
  steps = 1000,
  k = 2,
  ...
)

Arguments

object: A fitted HubbellGLM object.
scope: A formula or list of formulas defining the range of models to consider. See step for details.
scale: Scaling factor for the AIC penalty. Default is 0.
direction: Direction of stepwise search: "both", "backward", or "forward".
trace: Integer; if positive, prints information at each step.
keep: A function to retain extra information at each step.
steps: Maximum number of steps. Default is 1000.
k: Penalty multiplier for the number of parameters (2 for AIC).
...: Additional arguments passed to the fitting function.