Discussion List Archives

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Expressions for R factors

  • To: Multiple recipients of list <coredmg@iucr.org>
  • Subject: Expressions for R factors
  • From: Brian McMahon <bm@iucr.org>
  • Date: Fri, 4 Sep 1998 11:44:39 +0100 (BST)
John Bollinger at Indiana writes:

> With regard to _refine_ls_wR_factor_*: it seems unusual to me that the
> sum in the denominator is over Y(calc) instead of over Y(obs).  Is this
> intentional, and if so, can you refer me to any literature about why this
> particular usage is to be preferred?  I have never run into it before, and I
> shall have to make special provisions to get my software to calculate it.  I
> see that this formula appears in the version 2.0.1 dictionary as well, but
> that does not invalidate the question.

For example, the definition of _refine_ls_wR_factor_all is

                    ( sum | w | Y(obs) - Y(calc) |^2^ | )^1/2^
               wR = ( --------------------------------- )
                    (         sum | w Y(calc)^2^ |      )

as against the definition for _refine_ls_R_factor_gt, which is

                   sum | F(obs) - F(calc) |
               R = ------------------------
                         sum | F(obs) |

I note that the original definition for _refine_ls_wR_factor_ in version 1.0
of the Core read 

               wR = [sum(w|Ym-Yc|^2^) / sum(wYm^2^)]^1/2^

Have we simply introduced a typo into later versions?

Brian

[Send comment to list secretary]
[Reply to list (subscribers only)]