Constrained optimization via quadratic programming generated nonnegative weights minimizing score distortion while enforcing equal subgroup influence, unlike regression case-mix adjustment or post hoc ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results