relation of fit range with delta2

[ali dordaee]

 Hi every one,

 I hope all is well. I was fitting, I saw really differet amounts for delta2. For instace, we I was fitting from 0-20 Angstroms, delta2 was (for example) 8, and the second time when I tried to fit from 0-21 Angstroms, delta2 was 3.5, and the next time, when I was fitting from 0-19 Angstroms, the obtained delta2 was  -4 ! 

 The numbers are not exact, but I have seen such behaviors from diffpy-cmi when I was changing the fit range only by 1 Angstrom each time. Could anybody please disclose this to me? Thanks in advance.

[Simon Billinge]

Thanks Ali,

When things like this happen it is good to check the variable correlations.  Is the variable that handles delta2 correlated strongly with any other parameter?

--
You received this message because you are subscribed to the Google Groups "diffpy-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diffpy-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/diffpy-users/2d722cfc-ea9b-4834-9218-3469666149ban%40googlegroups.com.

--

Simon Billinge

Professor, Columbia University

Physicist, Brookhaven National Laboratory

[ali dordaee]

Yes Simon, it is correlated with Psize which is my particle size, but the thing I have also seen in diffpy sample codes is even those sample codes have one or two warnings of correlations. By the way, I don't know why the threshold is taken as 0.5, why not 0.6? I'd appreciate it if you could please address some resources in this regard.

 Thanks in advance,

 Bests,