CCL: The use of NMR spectra computations for verification of computational method

 Sent to CCL by: Grigoriy Zhurko [reg_zhurko^chemcraftprog.com]
 I have been computing NMR spectra of several organic compounds, comparing them
 with the experiment. Usually these computations show big systematic errors, but
 correlate well with the experiment (for  C13 spectra of organic molecules, I got
 the correlation coefficient R about 0.9995 with B3LYP/6-311G(D,P) method).
 Now I have computed some NMR spectra of bilirubin molecule with different DFT
 functionals, and I found that the correlation coefficient is not a good criteria
 of computation accuracy in my case. This molecule has internal hydrogen bonds,
 and different functionals (in particular, B3LYP and PBE) give quite different
 O..H distance (the difference is about 0.1 A), while other bond lengths in this
 molecule do not differ significantly (the difference is 0.012 A or less). The
 PMR spectrum with B3LYP correlates with the experimental one with R=0.997, and
 with PBE ? R=0.995. These values do not differ very much. But the coefficient B
 in the equation Y=A+B*X (for the linear approximation of experiment vs theory
 graph) is 1.02 for B3LYP, and 1.14 for PBE. So, with PBE it is far less from 1.
 Does that mean that PBE is much less appropriate method for this task?
 I suppose, that the systematic error of absolute values of the NMR chemical
 shifts is caused by unclear physical meaning of these chemical shifts and
 shieldings (maybe the solvent gives some additional shielding in experiment).
 So, my question is, whether the B coefficient in correlation must be always
 equal 1. If yes, then instead of correlation coefficients R I should use another
 criteria of computation accuracy ? the RMS of MAE difference between the
 computed and experimental chemical shifts, if the shielding of the standard
 (TMS) simply fitted for best agreement (not computed quantum-chemically). Is
 that correct? In my case, these MAE difference must be much bigger for PBE
 functional, than for the B3LYP functional.
 Grigoriy Zhurko.