Anyone recognize this single parameter discrete probability distribution?

Click For Summary
SUMMARY

The discussion centers on a single parameter discrete probability distribution characterized by the probability mass function (pmf) Pr(k;L) = L^k / (k! * k! * I_{0}(2*√L)), where I_{0}() denotes the modified Bessel function of the first kind. The user, J, seeks a closed-form solution for the distribution mean, noting that E(k²) = L and providing asymptotic behavior as L approaches infinity. J identifies the distribution as a special case of the Conway–Maxwell–Poisson distribution with the non-Poisson parameter ν = 2, emphasizing the need for an exact solution over asymptotic approximations.

PREREQUISITES
  • Understanding of discrete probability distributions
  • Familiarity with the modified Bessel function of the first kind
  • Knowledge of the Conway–Maxwell–Poisson distribution
  • Basic concepts of statistical moments, specifically mean and variance
NEXT STEPS
  • Research the properties and applications of the Conway–Maxwell–Poisson distribution
  • Study the modified Bessel function of the first kind and its implications in probability
  • Explore methods for deriving closed-form solutions for discrete distributions
  • Investigate asymptotic analysis techniques in probability theory
USEFUL FOR

Mathematicians, statisticians, and data scientists interested in advanced probability distributions, particularly those working with discrete models and seeking to derive exact solutions for complex distributions.

jacobcdf
Messages
5
Reaction score
0
I have a single parameter discrete probability distribution defined over the domain of non-negative integers with pmf in k of:

[tex]Pr(k;L) = \frac{L^{k}}{k! * k! * I_{0}(2*\sqrt{L})}[/tex]

Where [tex]I_{0}()[/tex] is the modified Bessel function of the first kind with order 0.

I do know that [tex]E(k^{2}) = L[/tex].

Can anyone come up with a closed form for the distribution mean?

Does anyone recognize this distribution?

Thanks in advance,
J.
 
Last edited:
Physics news on Phys.org
I also know that as [tex]L \rightarrow \infty[/tex]:

[tex]\gamma_{1} \rightarrow \sqrt{\frac{1}{2*\sqrt{K}}[/tex]

and

[tex]\gamma_{2} \rightarrow \gamma_{1}^{2}[/tex]

and E(k) appears to approach something approximated by:

[tex]\sqrt{L - \frac{\sqrt{L}}{2}}[/tex]

But regardless, I still would like an exact closed-form solution, as the asymptotic approximation appears of little use practically.


J.
 
In case anyone's interested, this distribution appears to be a special case of the Conway–Maxwell–Poisson distribution with the non-Poisson parameter [tex]\nu = 2[/tex].
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 25 ·
Replies
25
Views
3K
  • · Replies 14 ·
Replies
14
Views
6K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 5 ·
Replies
5
Views
1K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 24 ·
Replies
24
Views
7K
  • · Replies 7 ·
Replies
7
Views
2K