mabied
Dear all
Hi,
My problem already solved, just for sharing the answer with you!
"I used Rietveld refinement using rigid body of powder data, and although using do-error and OUT_CIF_Str there is no e.s.d included in out file or CIF, do you know how I can do it?
I want to include e.s.d of the atomic positions in cif file to have some thing like this _atom_site_label
_atom_site_fract_x
_atom_site_fract_y
_atom_site_fract_z
C1 0.1009(3) 0.57943(14) 0.3038(3)"
Alan Coelho, thankfully, sent me the following answer
The problem is that rigid body parameter errors does not get transferred to fractional coordinates in Version 4.1 and prior. Version 5 to be released soon does propagate the errors.
The work around for Version 4.1 is to use the boot strap method of error determination (Efron & Tibshirani 1986, DiCiccio & Efron 1996, Chernick 1999), ie. bootstrap_errors 200
I have previously played a bit with boot strapping (as it's called) and the errors reported seems to be the same as A matrix inversion; it does not give correlations however and is typically suited for very large problems where it's impractical to invert the A matrix.
Cheers
Alan