On 3/4/14, 11:37 AM, Evan Ward wrote:
> Phil suggested we discuss the relationship of the two least squares
> packages in [math]
>
> Current Status
>
> Currently both fitting.leastsquares and stat.regression have least
> squares implementations. The fitting.leastsquares package supports
> nonlinear least squares with weights, pluggable optimization
> algorithms, and (soon) data editing. The package is written from an
> optimization/engineer's perspective.
>
> As far as I can tell the stat.regression implementations are for linear
> least squares. Some of the implementations contain neat optimizations,
> for example requiring O(1) space for n data points. This package seems
> to be written more from a statistician/economist's perspective.
>
> Options
>
> 1. Keep separate packages.
> 2. delegate the implementation from one package to the other
> 3. merge into a single package. (could lead to some interesting
> algorithms. e.g. nonlinear general least squares)
>
> Phis, please add any important points I've missed.
This is correct. I don't see a big problem here  it is just worth
asking where the residual analysis / data editing functionality
belongs  in both places?
Also, it is worth considering merging the implementations. In the
stats classes, these would be the nonupdating impls (those that
require the full set of data in memory). It may well be that the
only material commonality is what is already exploited  the .linear
decompositions. Residual analysis may be something that can be shared.
Phil
>
> Best Regards,
> Evan
>
>> Phil Steitz commented on MATH1105:
>> 
>>
>> Might be better to take this discussion to the ML. We now have two
>> least squares impls  one in fitting/leastsquares and another in
>> stats.regression (actually this has been true for some time). The
>> stats side of it (residual analysis, ANOVA, etc.) belongs more
>> naturally in stats.regression. It might make more sense to add this
>> functionality there. Or maybe we just refactor to have the
>> stats.regression classes use the impl in leastsquares. In any case, we
>> should discuss on the ML.
>>
>>> Least squares statistical data editing
>>> 
>>>
>>> Key: MATH1105
>>> URL: https://issues.apache.org/jira/browse/MATH1105
>>> Project: Commons Math
>>> Issue Type: Improvement
>>> Reporter: Evan Ward
>>> Attachments: 0001Addstatisticaleditingcapability.patch,
>>> 0002IntegratedataeditingwiththeLSframework.patch
>>>
>
>
>
> 
> To unsubscribe, email: devunsubscribe@commons.apache.org
> For additional commands, email: devhelp@commons.apache.org
>
>

To unsubscribe, email: devunsubscribe@commons.apache.org
For additional commands, email: devhelp@commons.apache.org
