Open fscottfoti opened 10 years ago
Might want to add another item to this list - if the chi-squared check isn't sufficient, perhaps we should just run the sanity check of computing the marginals on the synthesized population and see how they compare to the original. The chi-squared check might cover this though.
Looks like SciPy has a chi-squared function: http://docs.scipy.org/doc/scipy-0.14.0/reference/generated/scipy.stats.chisquare.html
Is this done at this point?
We probably need an open issue on drawing methods.
Any other drawing methods? I tend to think doing the one from the paper will be sufficient.