After making changes to the code and/or documentation, this page should remain on the wiki as a reminder of what was done and how it was done. However, there is no guarantee that this page is updated in the end to reflect the final state of the project
So chances are that this page is considerably outdated and irrelevant. The notes here might not reflect the current state of the code, and you should not use this as serious documentation.
Explain how to create cfg.design for ft_*statistics
cfg.design as input into ft_*statistics (freq, timelock, or source) should be designed as follows: size MxN, where N is the number of trials/conditions
M is variable length. Along with cfg.design, the other options cfg.ivar, cfg.uvar, cfg.wvar, and cfg.cvar may also need to be specified. They are an index (indices) into which row(s) of the cfg.design correspond to 'independent variables', 'units of observation', 'within-block variables' and 'control variables', respectively.
Not all are needed: see statistics_*.m for what vectors it may include.
See http://fieldtrip.fcdonders.nl/walkthrough#non-paired_comparison and http://fieldtrip.fcdonders.nl/walkthrough#paired_comparison for examples on ivar and uvar. and http://fieldtrip.fcdonders.nl/development/statistics for uvar.
Is there any explanation on the wiki for more on uvar and wvar? (Seems it is all on the email discussion list specific to people's questions).
links to already existing pages with some mention of design matrix
Hierarchy of functions
(figure/diagram helpful http://fieldtrip.fcdonders.nl/development/statistics)
User calls ft_*statistics (freq, timelock, or source)
cfg.method: different methods for calculating probability of null hypothesis will call function: statistics_(cfg.method)
however, in ft_sourcestatistics, instead sourcestatistics_(cfg.method) is called, or statistics_wrapper if no method specified.
statistics_montecarlo.m (only) calls resampledesign.m
statistics_stat.m (deprecated: see http://fieldtrip.fcdonders.nl/example/statistics_toolbox )