The following tools are implemented as modular components to be used by any data analysis technique:
- Numerous high-precision optimisation algorithms.
- Model selection (Chen et al., 2004; d'Auvergne and Gooley, 2003):
- Akaike's Information Criteria (AIC).
- Small sample size corrected AIC (AICc).
- Bayesian or Schwarz Information Criteria (BIC).
- Bootstrap model selection.
- Single-item-out cross-validation (CV).
- Hypothesis testing ANOVA model selection (only the model-free specific technique of Mandel et al. (1995) is supported).
- Monte Carlo simulations (error analysis for all data analysis techniques).
- Model elimination - the removal of failed models prior to model selection (d'Auvergne and Gooley, 2006).
The relax user manual (PDF), created 2020-08-26.