On Sun, 20 Feb 2005 06:20:54 -0800 (PST), Leonid Solovyov wrote:

>
>> with DDM the deviations between known and estimated content
>> are twice larger compared to Rietveld.
>
>In this case the problem might be not in the DDM method, but in its
>particular realization in BGMN. The most probable reason is the minimal
>convolution intervals used. For noisy data the interval choice is
>essential.

I have set online BGMN level 3.5.3, which uses variable
convolution intervals as proposed by Leonid.

Regards,

Joerg Bergmann, Dresden
*** This EComStation system uptime is 0 days 2 hours ***

Reply via email to