This is very similar to the paper http://www.isi.edu/~szekely/contents/papers/2012/szekely2012-iui.pdf http://www.isi.edu/~szekely/contents/papers/2012/szekely2

This is very similar to the paper http://www.isi.edu/~szekely/contents/papers/2012/szekely2012-iui.pdf http://www.isi.edu/~szekely/contents/papers/2012/szekely2012... As per the paper , you can choose arbitrary distributions , construct a fluent graph , run Monte Carlo simulation and get the result - |via http://bit.ly/hnbuzz01 http://bit.ly/hnbuzz01 |