commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Phil Steitz <phil.ste...@gmail.com>
Subject Re: inverseCumAccuracy is probably not necessary
Date Sun, 04 Aug 2013 16:42:12 GMT
On 8/4/13 7:44 AM, Ajo Fod wrote:
> Guys,
>
> What is the use of inverseCumAccuracy when people want to instantiate an
> AbstractRealDistribution with a random generator?
> org.apache.commons.math3.distribution.AbstractRealDistribution<http://commons.apache.org/proper/commons-math/apidocs/org/apache/commons/math3/distribution/AbstractRealDistribution.html>
>
> I personally only seem to need to instantiate these objects with a
> RandomGenerator but never with the inverseCumAccuracy set to anything but
> DEFAULT_INVERSE_ABSOLUTE_ACCURACY.
>
> Could we be better of with inverseCumAccuracy moved to another constructor?

Good point.  We should take a careful look at which implementations
actually use this parameter and remove it from constructors for
those that don't.  I think some used to use it, but do not any
longer - in particular the ones that no longer rely on the default
inverse cum provided by AbstractRealDistribution.  The default impl
uses a solver to directly invert the cdf and that is what this
parameter is used for.  For the ones that do use it, it would also
be convenient to provide constructors that include distribution
parameters + random generator without this parameter.

Phil
>
> Thanks,
> Ajo.
>


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org


Mime
View raw message