If Koopman's theorem is not valid for dft calculations, how can we calculate global reactivity parameters from the HOMO-LUMO energies. Is there any correction factors for better result?
This has to do with the non-exact nature of the v_xc, as suggested by the work of Baerends and Gritsenko (and others).
E.g:
The analog of Koopmans’ theorem for virtual Kohn–Sham orbital energies (2009, http://www.ingentaconnect.com/content/nrc/cjc/2009/00000087/00000010/art00014)
The Kohn–Sham gap, the fundamental gap and the optical gap: the physical meaning of occupied and virtual Kohn–Sham orbital energies (2013, http://pubs.rsc.org/en/Content/ArticleLanding/2013/CP/C3CP52547C#!divAbstract)
Physical Meaning of Virtual Kohn–Sham Orbitals and Orbital Energies: An Ideal Basis for the Description of Molecular Excitations (2014, http://pubs.acs.org/doi/abs/10.1021/ct500727c)