[
https://issues.apache.org/jira/browse/MATH-1491?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16867867#comment-16867867
]
Virendra Singh commented on MATH-1491:
--------------------------------------
Hi [~lingchao], I looked into your issue, I checked it on version 2.2 & 3.6
with your Test case. I found that version 2.2 gives the Percentile value =
*7.55* while the version 3.6 gives the Percentile value =
*7.550000000000001*(for this particular test case).
It can be solved if you add delta/tolerance *1e-15* in
assertEquals(expected,actual,delta) method.
> Percentile computational accuracy issue
> ---------------------------------------
>
> Key: MATH-1491
> URL: https://issues.apache.org/jira/browse/MATH-1491
> Project: Commons Math
> Issue Type: Bug
> Affects Versions: 3.4, 3.4.1, 3.5, 3.6, 3.6.1
> Environment: System: Linux testinglab 4.4.0-131-generic
> #157~14.04.1-Ubuntu
> Java version "1.8.0_191"
> Java(TM) SE Runtime Environment (build 1.8.0_191-b12)
> Java HotSpot(TM) 64-Bit Server VM (build 25.191-b12, mixed mode)
>
>
> Reporter: Lingchao Chen
> Priority: Major
> Labels: performance
> Attachments: BugDemo.java
>
>
> Hi,
> The percentile method works well on the older versions, e.g., the version
> before 3.4. However, when I update commons-math to the newer version, there
> produces a computational accuracy issue. There is a backward compatibility
> bug behind it.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
