Re: Spectrum data model: accuracy

From: Alberto Micol <Alberto.Micol-at-eso.org>
Date: Fri, 15 Sep 2006 19:24:40 +0200

On Sep 15, 2006, at 19:17, Anita Richards wrote:

>>
>> Just to clarify that I meant not to add errors to bounds;
>> instead in this case I think that the Location.Value is 1975,
>> and the Location Error is 10 years.
>
> Sorry, that is also different from our intentions at least in Char
> - the top levels of description are inevitably imprecise because
> they are simple, so they do not have errors attatched.
>
> Mayeb this will become clearer next week...
> cheers
> a

My fault... I really meant:

Spectrum.Char.TimeAxis.Accuracy.SysError = 10 years

What is important is that the Bounds have nothing to do with the precision by which we know the location; the SysError and the StatError should be used to describe that.

The typical Bounds values will be the actual min and max of the *known* data values.

A nice weekend,

Alberto Received on 2006-09-15Z19:25:02