[time-nuts] ADEV vs. OADEV

Magnus Danielson magnus at rubidium.dyndns.org
Fri Jan 23 02:26:49 UTC 2009


Bruce,

> Tom
> 
> Its important to realise these calculated statistics are all estimators
> of the underlying Allan variance.
> Thus it is important to assign unique unambiguous names  to each of them
> so that some idea of their region of reasonable accuracy and perhaps
> their confidence limits can be estimated.

It should be recalled that AVAR is actually not necessarilly a name as 
such but short-hand. To help making the issue complex, AVAR has been 
given as a name to denote a particular algorithm, just as MADEV, TDEV 
etc. Even if one can argue that they have these meanings within a 
specific context, the ambiguity problem is there as it makes it harder 
to compare results.

> Thus for example:
> AVAR doesn't actually denote the underlying Allan variance but rather a
> particular algorithm used to estimate it from the data.
> 
> OAVAR denotes another algorithm for estimating the underlying Allan
> variance from the data

Trouble is, the algorithm Tom and Ulrich wants to denote OAVAR others 
have already denoted AVAR, thus causing ambiguity.

> TOTVAR, THEO1 and THEOH denote other algorithms that can be used to
> estimate the underlying Allan variance from the data.
> 
> Some of these algorithms produce biased estimates of the underlying
> Allan variance so correction for such bias is usually necessary.

The way the confidence interval closes on the estimate should be of a 
concern. We can calculate whatever we want, but as we try to judge the 
results of such estimates we should also view the indicators of its 
precision if available.

Cheers,
Magnus




More information about the Time-nuts_lists.febo.com mailing list