[
https://issues.apache.org/jira/browse/RNG-105?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alex D Herbert closed RNG-105.
------------------------------
Resolution: Duplicate
Duplicate of RNG-103.
This issue was previously created. Sorry.
> Use the KempSmallMeanPoissonSampler in the LargeMeanPoissonSampler
> ------------------------------------------------------------------
>
> Key: RNG-105
> URL: https://issues.apache.org/jira/browse/RNG-105
> Project: Commons RNG
> Issue Type: Improvement
> Components: sampling
> Affects Versions: 1.3
> Reporter: Alex D Herbert
> Assignee: Alex D Herbert
> Priority: Trivial
> Fix For: 1.3
>
>
> The LargeMeanPoissonSampler requires a Poisson sample using a mean less than
> 1. This is the fractional component of the large mean.
> The KempSmallMeanPoissonSampler outperforms the SmallMeanPoissonSampler when
> the mean is less than 1. This can be significant if the underlying generator
> is slow. The Kemp sampler only requires a single uniform deviate.
> Switch the algorithm used in the LargeMeanPoissonSampler.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
