R-factor

In GED two types of R-factors are in use: structural R-factor (R_str, sometimes called just R-factor) and experimental R-factor (R_exp).

Structural R-factor is a measure of the agreement between the model and the experimental molecular intensity curves, which are traditionally defined either as sM(s) or s⁴I_m(s).

In case of using sM(s) molecular intensity curves the structural R-factor in percent units is defined as

<math>R_{str} = \sqrt {\frac{\sum_{i}^{N} \sum_{j}^{M} (sM(s)^{m}_{i,j} - sM(s)^{e}_{i,j})^2}{\sum_{i}^N \sum_{j}^M (sM(s)^{e}_{i,j})^2}} \times 100</math>

where sM(s)^m and sM(s)^e are model and experimental molecular intensity curves and summations are performed over all data points of all sM(s) curves. The smaller R_str the better agreement between model and experimental data. R_str usually ranges between 1.0 (small molecules and high quality data) and 10.0% (large molecules or low quality data) and depends on the amount of experimental data.

R_str is equal to R_D defined in <bib id="MastersSC2013" /> since it does not account for correlations between values in experimental data.