This repository contains the functions used to test the statistical functions in LIMO tools.
Two types of tests peformed:
- unit tests which runs the code base against ground truth data (ie know expected results)
- simulations and data driven approach which test peformances ensuring LIMO does what it is supposed to do to control outliers, error rates, etc
Read more on testing results on the wiki
Defining outlier trials
Weighting Scheme Comparison
IRLS performances
Bootrapping under the null