Ah, I have been ruminating about your theory here, and I realize now, after isolating the specific points of antipathy between my intrinsic experiental extrapolations and the supposed implications of your espousal using a holistic overview, what is needed to reconcile your formulas to empirical datum.
First, the parabolic. Then keep in mind Ulricks post. Although, Ulrick, I think that using the neo-classical Von Neumann/Morgenstern risk axioms, in this situation, is a bit simplistic.
Don't you think?
Look at Dirac's postulate of non-communicativitiness. That seems to keep the divergencess at the singularities down to a nugatory level.
Applied to the 4-th dimensional multiplicital-derivative (that is e to the integral of the log of the function from a to b, divided by b minus a), this conjecture seems to imply a double convexity of the topological hull of the projected data, sufficient to converge the 2nd order derivatives to unity, while refracting the 3rd orders at these points to infinity. (Which is what all girls want! Duh!)
So, using your modulated theory, I ran a simulation on my associates supercomputer, and we came up with the constant for women:
W=01010011011010110110010101100101011101000010110000100000011100110110101101100101011001010111010000101100001000000111001101101011011001010110010101110100001011000010000001110011011010110110010101100101011101000010110000100000011100110110101101100101011001010111010000100001
Pardon the binary, but since man and women can be viewed as positive and negative entities interacting in a discontinuous, yet paridoxically enclosed environment, we felt it was justified.