The SwE toolbox for longitudinal and repeated measures neuroimaging data uses a marginal model with the sandwich estimator of standard errors. It is implemented as a Matlab toolbox for SPM8 or SPM12. Please contact Bryan Guillaume & Tom Nichols with any questions or issues.

Download

• Software:  SwE toolbox v1.2.5 (Last updated: 1st September 2016)
• Or, check GitHub for history and latest version: SwE Repo

Documentation

• Manual. (Please note this only covers parametric inference, and not the newly released Wild Bootstrap nonparametric inference).
  
• Worked example of parametric inference on a real repeated measures fMRI data.
• References
  • Fast and accurate modelling of longitudinal and repeated measures neuroimaging data, NeuroImage 94:2870302, 2014.
    This is the original paper, covering parametric inference and basic small sample adjustments
  • New small sample adjustments and the Wild Bootstrap are described in:OHBM2015 poster, thesis and paper (in preparation).

What's new?

• As of September 2015, better small sample adjustments have been added as well as the Wild Bootstrap to make non-parametric inferences with the Sandwich Estimator.
• (A paper covering these new developments is in preparations; please see Dr. Guillaume's thesis for full details.)

Why use SwE?

This approach has a number of advantages over traditional linear mixed effect models:

• Easy random effects
  Only the population model is specified, meaning that no random effects (e.g. random slopes) need to be specified.
  • Despite having no explicit specification, all possible random effects are accounted for through the use of an unstructured error covariance.
  • For moderate sample sizes, the "Hom"ogeneous sandwich estimator assumes each subject in a group shares the same visit-based covariance strucutre.
  • For large sample sizes, the "Het"ergoeneous (tranditional) sandwhich estimator doesn't even assume common covariance over subjects.
  • When comparing multiple groups, each group automatically has its own covariance structure.
• No convergence problems
  The population model is estimated with Ordinary Least Squares, meaning that the method is non-iterative and thus is immune to convergence failures not uncommon in complex mixed effects models.
• Built for neuroimaging
  • Toolbox for widely used SPM software
  • While the traditional sandwich estimator techniques often assume large samples, SwE implements carefully evaluated (and in some cases novel) degrees of freedom estimator and small sample adjustments.
  • Familywise error-corrected voxel and cluster inferences are available with the Wild Boostrap, avoiding any parametric (e.g. random field theory) distributional assumptions.

Bugs & Feedback

Please be sure to report any problems or questions to Bryan Guillaume & Tom Nichols.