You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@commons.apache.org by Will Glass-Husain <wg...@forio.com> on 2008/04/27 00:07:58 UTC
[math] normal distribution - probability mass function?
Hi,
I'm looking to include the capability in a project to calculate both the
cumulative normal distribution as well as the normal probability mass
function. (similar to Excel's NORMDIST function).
I see that NormalDistribution has a method to return the cumulative
probability, but does not have the probability function itself. Can anyone
comment on why this missing? (or if there's an alternative way to calculate
what I'm looking for?)
Thanks in advance,
WILL
Re: [math] normal distribution - probability mass function?
Posted by Phil Steitz <ph...@gmail.com>.
On Sat, Apr 26, 2008 at 5:18 PM, Brent Worden <br...@gmail.com> wrote:
> The pdf could definitely be used as a kernel for kernel density
> estimation.
>
Yes, I guess that's one use for at least the Gaussian pdf. So the
question is does it make sense to go ahead and add the pdf to the
continuoous distribution interface in 2.0? Should be doable for what
we have now and not too onerous for others, no?
Phil
> Brent
>
>
>
> -----Original Message-----
> From: Will Glass-Husain [mailto:wglass@forio.com]
> Sent: Saturday, April 26, 2008 6:59 PM
> To: Commons Users List
> Subject: Re: [math] normal distribution - probability mass function?
>
> Hi Phil,
>
> We offer an online simulation toolkit, which (among other things) offers
> a
> set of mathematical functions to users. ( forio.com/broadcast ). One
> of
> our users requested we add an Excel-like NORMDIST function for
> cumulative
> probability distribution. Since we are adding this function, I thought
> it'd
> be a good idea to offer a more complete set of distribution functions,
> including the probability mass function.
>
> Best, WILL
>
> > We can talk about this. Out of curiousity, what are you using this
> > (the pdf) for?
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
> For additional commands, e-mail: user-help@commons.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
For additional commands, e-mail: user-help@commons.apache.org
RE: [math] normal distribution - probability mass function?
Posted by Brent Worden <br...@gmail.com>.
The pdf could definitely be used as a kernel for kernel density
estimation.
Brent
-----Original Message-----
From: Will Glass-Husain [mailto:wglass@forio.com]
Sent: Saturday, April 26, 2008 6:59 PM
To: Commons Users List
Subject: Re: [math] normal distribution - probability mass function?
Hi Phil,
We offer an online simulation toolkit, which (among other things) offers
a
set of mathematical functions to users. ( forio.com/broadcast ). One
of
our users requested we add an Excel-like NORMDIST function for
cumulative
probability distribution. Since we are adding this function, I thought
it'd
be a good idea to offer a more complete set of distribution functions,
including the probability mass function.
Best, WILL
> We can talk about this. Out of curiousity, what are you using this
> (the pdf) for?
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
For additional commands, e-mail: user-help@commons.apache.org
Re: [math] normal distribution - probability mass function?
Posted by Will Glass-Husain <wg...@forio.com>.
Hi Phil,
We offer an online simulation toolkit, which (among other things) offers a
set of mathematical functions to users. ( forio.com/broadcast ). One of
our users requested we add an Excel-like NORMDIST function for cumulative
probability distribution. Since we are adding this function, I thought it'd
be a good idea to offer a more complete set of distribution functions,
including the probability mass function.
Best, WILL
> We can talk about this. Out of curiousity, what are you using this
> (the pdf) for?
Re: [math] normal distribution - probability mass function?
Posted by Phil Steitz <ph...@gmail.com>.
On Sat, Apr 26, 2008 at 4:01 PM, Will Glass-Husain <wg...@forio.com> wrote:
> Hi Phil,
>
> That makes sense -- Thanks!
>
> I used this formula in Excel and did a comparison with Excel-- it matched
> exactly (at 5 decimal places). Maybe it would make sense to add it in, just
> for comprehensiveness? But not a big deal.
We can talk about this. Out of curiousity, what are you using this
(the pdf) for?
>
> I don't think I've posted with a Math question before, but as long as I'm
> here-- to me Commons-Math is one one of the unsung niche heroes of Apache.
> Serves a specific need, nicely documented, works well. I use it extensively
> in a particular app, but rarely have questions about it since it just
> works. Thanks to you and the other authors for creating it.
Thanks!
Phil
>
> Cheers, WILL
>
>
>
>
>
> On Sat, Apr 26, 2008 at 3:46 PM, Phil Steitz <ph...@gmail.com> wrote:
>
> > On Sat, Apr 26, 2008 at 3:07 PM, Will Glass-Husain <wg...@forio.com>
> > wrote:
> > > Hi,
> > >
> > > I'm looking to include the capability in a project to calculate both
> > the
> > > cumulative normal distribution as well as the normal probability mass
> > > function. (similar to Excel's NORMDIST function).
> > >
> > > I see that NormalDistribution has a method to return the cumulative
> > > probability, but does not have the probability function itself. Can
> > anyone
> > > comment on why this missing? (or if there's an alternative way to
> > calculate
> > > what I'm looking for?)
> > >
> > > Thanks in advance,
> > >
> > > WILL
> > >
> >
> > Commons-math does not currently provide numerical estimates of
> > probability density functions for continuous distributions. I guess
> > this is because most applications need just need probabilities or
> > quantiles. I am not sure how good the numerics of the definitional
> > formula for the normal pdf is, but you could just implement that
> > directly. See, e.g.
> > http://mathworld.wolfram.com/GaussianFunction.html.
> >
> > Phil
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
> > For additional commands, e-mail: user-help@commons.apache.org
> >
> >
>
>
> --
> Forio Business Simulations
>
> Will Glass-Husain
> 415 440-7500x89
> wglass@forio.com
> www.forio.com
>
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
For additional commands, e-mail: user-help@commons.apache.org
Re: [math] normal distribution - probability mass function?
Posted by Will Glass-Husain <wg...@forio.com>.
Hi Phil,
That makes sense -- Thanks!
I used this formula in Excel and did a comparison with Excel-- it matched
exactly (at 5 decimal places). Maybe it would make sense to add it in, just
for comprehensiveness? But not a big deal.
I don't think I've posted with a Math question before, but as long as I'm
here-- to me Commons-Math is one one of the unsung niche heroes of Apache.
Serves a specific need, nicely documented, works well. I use it extensively
in a particular app, but rarely have questions about it since it just
works. Thanks to you and the other authors for creating it.
Cheers, WILL
On Sat, Apr 26, 2008 at 3:46 PM, Phil Steitz <ph...@gmail.com> wrote:
> On Sat, Apr 26, 2008 at 3:07 PM, Will Glass-Husain <wg...@forio.com>
> wrote:
> > Hi,
> >
> > I'm looking to include the capability in a project to calculate both
> the
> > cumulative normal distribution as well as the normal probability mass
> > function. (similar to Excel's NORMDIST function).
> >
> > I see that NormalDistribution has a method to return the cumulative
> > probability, but does not have the probability function itself. Can
> anyone
> > comment on why this missing? (or if there's an alternative way to
> calculate
> > what I'm looking for?)
> >
> > Thanks in advance,
> >
> > WILL
> >
>
> Commons-math does not currently provide numerical estimates of
> probability density functions for continuous distributions. I guess
> this is because most applications need just need probabilities or
> quantiles. I am not sure how good the numerics of the definitional
> formula for the normal pdf is, but you could just implement that
> directly. See, e.g.
> http://mathworld.wolfram.com/GaussianFunction.html.
>
> Phil
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
> For additional commands, e-mail: user-help@commons.apache.org
>
>
--
Forio Business Simulations
Will Glass-Husain
415 440-7500x89
wglass@forio.com
www.forio.com
Re: [math] normal distribution - probability mass function?
Posted by Phil Steitz <ph...@gmail.com>.
On Sat, Apr 26, 2008 at 3:07 PM, Will Glass-Husain <wg...@forio.com> wrote:
> Hi,
>
> I'm looking to include the capability in a project to calculate both the
> cumulative normal distribution as well as the normal probability mass
> function. (similar to Excel's NORMDIST function).
>
> I see that NormalDistribution has a method to return the cumulative
> probability, but does not have the probability function itself. Can anyone
> comment on why this missing? (or if there's an alternative way to calculate
> what I'm looking for?)
>
> Thanks in advance,
>
> WILL
>
Commons-math does not currently provide numerical estimates of
probability density functions for continuous distributions. I guess
this is because most applications need just need probabilities or
quantiles. I am not sure how good the numerics of the definitional
formula for the normal pdf is, but you could just implement that
directly. See, e.g.
http://mathworld.wolfram.com/GaussianFunction.html.
Phil
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
For additional commands, e-mail: user-help@commons.apache.org