[ https://issues.apache.org/jira/browse/MATH1203?page=com.atlassian.jira.plugin.system.issuetabpanels:commenttabpanel&focusedCommentId=14354888#comment14354888
]
Phil Steitz edited comment on MATH1203 at 3/10/15 1:43 PM:

Thanks for reviewing, Thomas. I did think about checking variance = 0 instead and probably
should have done that. In fact, since it is possible that a nonconstant usersupplied kernel
can have no support in a bin, the 0/0 computation that led to MATH1208 could happen even
without variance == 0. So I guess a more robust test is first check if the distribution variance
is 0, then check kB > 0. Thanks again for reviewing.
was (Author: psteitz):
Thanks for reviewing, Thomas. I did think about checking variance = 0 instead and probably
should have done that. In fact, since it is possible that a nonconstant usersupplied kernel
can have no support in a bin, the 0/0 computation that led to MATH1208 could happen even
without variance == 0. So I guess a more robust test is first check if the distribution is
variance is 0, then check kB > 0. Thanks again for reviewing.
> getKernel fails for buckets with only multiple instances of the same value in random.EmpiricalDistribution
> 
>
> Key: MATH1203
> URL: https://issues.apache.org/jira/browse/MATH1203
> Project: Commons Math
> Issue Type: Bug
> Affects Versions: 3.4, 3.4.1
> Reporter: Kyle Kavanagh
>
> After loading a set of values into an EmpericalDistribution, assume that there's a case
where a single bin ONLY contains multiple instances of the same value. In this case the standard
deviation will equal zero. This will fail when getKernel attempts to create a NormalDistribution.
The other case where stddev=0 is when there is only a single value in the bin, and this is
handled by returning a ConstantRealDistribution rather than a NormalDistrbution.
> See: https://issues.apache.org/jira/browse/MATH984

This message was sent by Atlassian JIRA
(v6.3.4#6332)
