commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sébastien Brisard <sebastien.bris...@m4x.org>
Subject Re: [Math] Question about "AbstractLeastSquaresOptimizer" code
Date Wed, 25 Apr 2012 15:56:32 GMT
2012/4/25 Gilles Sadowski <gilles@harfang.homelinux.org>:
> On Wed, Apr 25, 2012 at 03:46:58PM +0200, Sébastien Brisard wrote:
>> Hi Gillles,
>> >
>> > Can we then agree that the current naming is not really clear?
>> > Since we are not allowed to readily remove this method, the best solution
>> > would be to update the Javadoc to make it clear what is done there, i.e.
>> > cite the full definition of ASE (together with a reference).
>> >
>> this would be the best solution in my view. I just need to find a good
>> reference where ASE is properly defined.
>
> The ones you found are good enough, I guess; just to let people know that
> they might not want to use that method. ;-)
>
Oh no, they do want to use this method. From what I gathered, it's
actually the standard way of estimating the error on the parameters. I
found two nice references
1. http://www.itl.nist.gov/div898/strd/ (NIST is your friend -- will
implement some unit tests based on these reference cases!)
2. http://en.wikipedia.org/wiki/Least_squares see paragraph "In a
least squares calculation with unit weights, or in linear regression,
the variance on the jth parameter"

Best regards,
Sébastien


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org


Mime
View raw message