miscellaneous, mostly matlab/octave tools for statistical things
this function takes two columns of (un-paired) data and shuffles the data to create a particular difference between the two columns (i.e., a particular t-statistic) this could be a good tool for teaching, or for matching-up (small) samples of data; larger samples (N>10?) will suffer from a combinatorial explosion. unfortunately, this would be a good tool for creating a particular effect size given reasonable-looking data (i.e., for fraud)