commons-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Luc Maisonobe (JIRA)" <j...@apache.org>
Subject [jira] Resolved: (MATH-229) Add support for removal of observations from DescriptiveStatistics
Date Tue, 11 Nov 2008 17:53:44 GMT

     [ https://issues.apache.org/jira/browse/MATH-229?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Luc Maisonobe resolved MATH-229.
--------------------------------

    Resolution: Fixed

fixed as of r713093 (in MATH_2.0 branch).

The patch has been applied with some minor changes in tests (removed print statements, replaced
tab characters by spaces). The test of the removal feature by itself has been included in
the Junit tests
for DescriptiveStatistics.

Thanks for providing the patch.

> Add support for removal of observations from DescriptiveStatistics
> ------------------------------------------------------------------
>
>                 Key: MATH-229
>                 URL: https://issues.apache.org/jira/browse/MATH-229
>             Project: Commons Math
>          Issue Type: Improvement
>         Environment: Feature-related, platform-independent
>            Reporter: Cyril Briquet
>            Priority: Minor
>         Attachments: ObservationRemovalFeaturePatch, ObservationRemovalTest.java, ObservationRemovalTestPatch
>
>   Original Estimate: 2h
>  Remaining Estimate: 2h
>
> [Objective]
> Add support for removal of observations from DescriptiveStatistics.
> [Rationale]
> Typically, a process can be modeled through a set of values.
> These values can be fed to DescriptiveStatistics through the addValue() method.
> The user might be interested in recomputing e.g. the mean and variance
> of the observed values each time a new value is added.
> Sometimes, however, a value of the observed process
> takes some time to acquire. Nonetheless, the user might want
> to recompute stats both (1) with the initial estimation
> of the most recent value as well as (2) with each subsequent update
> of the the most recent value.
> [Example]
> The values 50 and 60 are added.
> Then the observed process leads to a new observation
> that has an initial value of 42.
> Then this most recent value stabilizes to a value of 40. 
> Stats should be computed on the sets of observations
> { 50, 60 }, { 50, 60, 42 } and { 50, 60, 40 }.
> [Requirements]
> Support should be added to both enable the update or removal
> of the most recent value added to a DescriptiveStatistics.
> The underlying data structure (i.e. ResizableDoubleArray)
> can be modified to this end.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message