Continuous binomial distribution: Difference between revisions

Declining submission: nn - Submission is about a topic not yet shown to meet general notability guidelines (be more specific if possible)
Cleanup: remove deletion notice and missing templates
 
(15 intermediate revisions by 6 users not shown)
Line 1: Line 1:
{{AFC submission|d|nn|u=Bayesianism|ns=118|decliner=Devonian Wombat|declinets=20260520060608|ts=20260226043005}} <!-- Do not remove this line! -->
{{AFC comment|1=Source 2 appears to be a calculator. [[User:Devonian Wombat|Devonian Wombat]] ([[User talk:Devonian Wombat|talk]]) 06:06, 20 May 2026 (UTC)}}
----
{{Short description|Continuous probability distribution on the unit interval}}
{{External links|date=February 2026}}
{{External links|date=February 2026}}


{{Use Oxford spelling |date=February 2026}}
{{Use Oxford spelling |date=February 2026}}
{{Use dmy dates |date=February 2026}}


{{Infobox probability distribution
{{Infobox probability distribution
Line 48: Line 40:
}}
}}


In [[probability theory]] and [[statistics]], the '''continuous binomial distribution''' (also called the '''cobin distribution''') is a family of continuous probability distributions on the unit interval that belongs to an [[exponential dispersion model|exponential dispersion family]]. It was introduced as a response distribution for [[generalized linear model]]s for continuous proportional data, proposed as an alternative to [[beta regression]].{{r|LeeDahlOvaskainenDunson2025}} The special case <math>\lambda=1</math> coincides with the [[continuous Bernoulli distribution]].
In [[probability theory]] and [[statistics]], the '''continuous binomial distribution''' (also called the '''cobin distribution''') is a family of continuous probability distributions on the unit interval that belongs to an [[exponential dispersion model|exponential dispersion family]]. The name ''continuous binomial'' (cobin) was adopted in 2026 in developing [[generalized linear model]]s for continuous proportional data, as an alternative to [[beta regression]] <ref name ="lee2026">{{Cite journal |last=Lee |first=Changwoo J. |last2=Dahl |first2=Benjamin K. |last3=Ovaskainen |first3=Otso |last4=Dunson |first4=David B. |date=2026-05-18 |title=Scalable and robust regression models for continuous proportional data |url=https://pmc.ncbi.nlm.nih.gov/articles/PMC13188389/ |journal=Journal of the American Statistical Association |doi=10.1080/01621459.2026.2626081 |issn=0162-1459 |pmc=13188389 |pmid=42169758}}</ref>.
The special case <math>\lambda=1</math> coincides with the [[continuous Bernoulli distribution]]<ref>{{Cite journal |last=Loaiza-Ganem |first=Gabriel |last2=Cunningham |first2=John |date=2019 |title=The continuous Bernoulli: fixing a pervasive error in variational autoencoders |url=https://proceedings.neurips.cc/paper/2019/hash/f82798ec8909d23e55679ee26bb26437-Abstract.html |journal=Advances in Neural Information Processing Systems |publisher=Curran Associates, Inc. |volume=32}}</ref>.


==Definition==
==Definition==
A random variable <math>X</math> is said to follow a continuous binomial (cobin) distribution with natural parameter <math>\theta</math> and inverse dispersion <math>\lambda\in\{1,2,\dots\}</math>, written <math>X\sim\mathrm{cobin}(\theta,\lambda^{-1})</math>, if it has density on <math>[0,1]</math> given by{{r|cobinCRANref}}
A random variable <math>X</math> is said to follow a continuous binomial (cobin) distribution with natural parameter <math>\theta</math> and inverse dispersion <math>\lambda\in\{1,2,\dots\}</math>, written <math>X\sim\mathrm{cobin}(\theta,\lambda^{-1})</math>, if it has density on <math>[0,1]</math> given by
:<math>
:<math>
f(x;\theta,\lambda)=h(x;\lambda)\exp\!\big(\lambda\theta x-\lambda B(\theta)\big),\qquad 0\le x\le 1,
f(x;\theta,\lambda)=h(x;\lambda)\exp\!\big(\lambda\theta x-\lambda B(\theta)\big),\qquad 0\le x\le 1,
Line 69: Line 62:
with <math>h(x;1)=1</math>. The function <math>h(x;\lambda)/\lambda</math> coincides with the probability density function of the [[Irwin–Hall distribution]] with parameter <math>n=\lambda</math>, evaluated at <math>\lambda x</math>.
with <math>h(x;1)=1</math>. The function <math>h(x;\lambda)/\lambda</math> coincides with the probability density function of the [[Irwin–Hall distribution]] with parameter <math>n=\lambda</math>, evaluated at <math>\lambda x</math>.


When <math>\lambda</math> is fixed, the cobin distribution belongs to a one-parameter [[natural exponential family]] in <math>\theta</math>.{{r|LeeDahlOvaskainenDunson2025}}
When <math>\lambda</math> is fixed, the cobin distribution belongs to a one-parameter [[natural exponential family]] in <math>\theta</math>.  
 
==Earlier studies==
 
The corresponding two-parameter distributional family appeared in earlier statistical literature<ref name="lee2026" /> without the name ''continuous binomial distribution''. For example, Bates (1955)<ref>{{cite journal |last=Bates |first=Grace E. |year=1955 |title=Joint Distributions of Time Intervals for the Occurrence of Successive Accidents in a Generalized Polya Scheme |journal=The Annals of Mathematical Statistics |volume=26 |issue=4 |pages=705–720 |doi=10.1214/aoms/1177728429}}</ref> studied the same distribution in the context of modeling time intervals between successive accidents and hypothesis testing against uniformity.


==Related distributions==
==Related distributions==
Line 83: Line 80:
\bar{X}=\frac{1}{\lambda}\sum_{i=1}^{\lambda} X_i
\bar{X}=\frac{1}{\lambda}\sum_{i=1}^{\lambda} X_i
\sim \mathrm{cobin}(\theta,\lambda^{-1}).
\sim \mathrm{cobin}(\theta,\lambda^{-1}).
</math>{{r|LeeDahlOvaskainenDunson2025}}
</math>


==Properties==
==Properties==


===Mean and variance===
===Mean and variance===
The mean and variance of <math>X\sim\mathrm{cobin}(\theta,\lambda^{-1})</math> can be expressed in terms of derivatives of <math>B(\theta)</math>:{{r|cobinCRANref}}
The mean and variance of <math>X\sim\mathrm{cobin}(\theta,\lambda^{-1})</math> can be expressed in terms of derivatives of <math>B(\theta)</math>:


* <math>\operatorname{E}(X)=B'(\theta)=\frac{e^\theta}{e^\theta-1}-\frac{1}{\theta}</math>, for <math>\theta\ne 0</math>.
* <math>\operatorname{E}(X)=B'(\theta)=\frac{e^\theta}{e^\theta-1}-\frac{1}{\theta}</math>, for <math>\theta\ne 0</math>.
* <math>\operatorname{Var}(X)=\frac{1}{\lambda}B''(\theta)=\frac{1}{\lambda}\left(\frac{1}{\theta^2}-\frac{e^\theta}{(e^\theta-1)^2}\right)</math>, for <math>\theta\ne 0</math>.
* <math>\operatorname{Var}(X)=\frac{1}{\lambda}B''(\theta)=\frac{1}{\lambda}\left(\frac{1}{\theta^2}-\frac{e^\theta}{(e^\theta-1)^2}\right)</math>, for <math>\theta\ne 0</math>.


If <math>\theta=0</math>, then <math>\operatorname{E}(X)=\frac{1}{2}</math> and <math>\operatorname{Var}(X)=\frac{1}{12\lambda}</math>.{{r|cobinCRANref}}
If <math>\theta=0</math>, then <math>\operatorname{E}(X)=\frac{1}{2}</math> and <math>\operatorname{Var}(X)=\frac{1}{12\lambda}</math>.


===Sufficient statistic for the mean===
===Sufficient statistic for the mean===
Line 111: Line 108:


==Applications==
==Applications==
The cobin distribution has been proposed as a response distribution for [[generalized linear model]]s of continuous proportional data, as an alternative to beta regression, including extensions with random effects.{{r|LeeDahlOvaskainenDunson2025}}
The cobin distribution has been proposed as a response distribution for [[generalized linear model]]s of continuous proportional data, as an alternative to beta regression, including extensions with random effects.


==References==
==References==
{{Reflist
{{reflist}}
|refs=
<ref name=LeeDahlOvaskainenDunson2025>
{{cite journal
|last1=Lee |first1=Changwoo J.
|last2=Dahl |first2=Benjamin K.
|last3=Ovaskainen |first3=Otso
|last4=Dunson |first4=David B.
|date=2026
|title=Scalable and robust regression models for continuous proportional data
|journal=Journal of the American Statistical Association
|pages=1–22
|doi=10.1080/01621459.2026.2626081
|arxiv=2504.15269
}}</ref><ref name=cobinCRANref>
{{citation
|title=cobin: Cobin and Micobin Regression Models for Continuous Proportional Data
|publisher=CRAN
|year=2025
|url=https://cran.r-project.org/web/packages/cobin/index.html
}}
</ref>
}}


{{ProbDistributions|continuous-bounded}}
{{ProbDistributions|continuous-bounded}}
[[:Category:Continuous distributions]]
[[Category:Continuous distributions]]
[[:Category:Exponential family distributions]]
[[Category:Exponential family distributions]]