Hi all,
Am I right in thinking there are roughly two camps in this dicussion?
Those who think that adding more data degrades the refinement if that data
is not useful and those who think it makes no difference. (I say 'not
useful' in the context of Vanadium in neutron data or deuterium in x-ray
data for examples).
My understanding is that the least squares matrix is made up of the
derivatives of the differences between data and model w.r.t the parameters
you are refining. So if a dataset has no useful information for a
parameter then the relevant derivatives are zero. So there should be no
degradation in the esd on the parameter, right? This would imply both
datasets should (always) be used simaltaneously.
I guess the degradation which is found would come from parameters which
are determined by both datasets and come out with different values in each
separate refinement. The higher esd reflects the disagreement between the
neutrons and x-rays. (Now enter the questions about precision and
accuracy.) Is this the area where there is a disagreement? What should one
do when the datasets disagree with each other? Ideally work out why and
model that in a combined fit! In practice either
-Do a combined fit and report postions with higher esd's (as you aren't
too sure where the atoms really are.)
-Do two separate fits, each having lower esd's but disagreeing with each
other.
Which is better?
Enough from me now, it's time to do some work today. Incidentally does
anyone have an example of a refinement where parameters are degraded by
the combined fit *and* they agree with each other when two separate fits
are carried out.
Jon Wright
PhD Student, Chemistry Dept, Cambridge Uni, UK.
On 25 May 1999, Andrew Wills wrote:
> Alan,
>
> I am not suggesting removing reflections. But, I think that we should make
> sure that we are combining the data in the best possible way. If we know have
> strong information on a vanadium position from X-rays and (extrapolate again)
> have only noise from neutrons, then stastically introducing the neutron data
> whilst no changing the best fit will degrade the least - squares approach to
> it. The final structure should fit all data, but are we approaching it
> optimally? I know that this is a can of worms, but it is good to think about
> what we are doing as combined refinements will continue to become less exotic.
>
> -Andrew
> --------------
> Andrew Wills
> Centre D'�tudes Nucl�aires de Grenoble
>
>
> "Alan Hewat, ILL Grenoble" <[EMAIL PROTECTED]> wrote:
> >If we have an atom that is seen by one
> >radiation and not by the other there will be a degradation in the quality of
> >the parameters by combining the refinement in the current fashion.
>
> Do you mean for example that we might degrade the parameters of a V atom
> by introducing neutron data ?
>
> I don't think this is true, but it is an interesting question. If we were to
> extrapolate this argument "ad absurdum" we could say that because some
> reflections (for a given radiation) do not give any information about some
> parameters (easy to demonstrate) then we would obtain better estimates
> for those parameters by removing those reflections from the least squares
> process. (Surely untrue :-)
