Configure an fMRI linear model — fmri_lm

fmri_lm_control() is the single statistical configuration boundary for fmri_lm(). Each section is a validated typed specification; mechanical choices such as chunking and parallelism belong in compute_spec().

Usage

fmri_lm_control(
  estimation = NULL,
  noise = NULL,
  robust = NULL,
  variance = NULL,
  weights = NULL,
  projection = NULL,
  na_action = c("error", "propagate"),
  ...
)

Arguments

estimation: An estimation_spec(). When omitted (NULL), the control object uses estimation_spec("runwise_meta"), the same default as fmri_lm(), so supplying a control object only to change another section (for example robust) does not change the estimation scope. Note that this differs from calling estimation_spec() with no arguments, which returns the joint scope. There are two exceptions: HAC, sandwich, or Satterthwaite inference requested through variance is implemented only for a joint fit, so an omitted estimation then resolves to estimation_spec("joint") with a message saying so; and with a registered engine, an omitted estimation resolves to estimation_spec("joint") (see fmri_lm()).
noise: A noise_spec().
robust: A robust_spec().
variance: A variance_spec().
weights: A weights_spec().
projection: A projection_spec().
na_action: Either "error" or "propagate".
...: Transitional legacy option lists retained for one compatibility window. New code should use the typed sections above.

Value

An object of class fmri_lm_control.