Negative multinomial distribution

In probability theory and statistics, the negative multinomial distribution is a generalization of the negative binomial distribution (NB(x0, p)) to more than two outcomes.[1]

Notation
Parameters — the number of failures before the experiment is stopped,
Rmm-vector of "success" probabilities,

p0 = 1 − (p1+…+pm) — the probability of a "failure".
Support
PMF
where Γ(x) is the Gamma function.
Mean
Variance
MGF
CF

As with the univariate negative binomial distribution, if the parameter is a positive integer, the negative multinomial distribution has an urn model interpretation. Suppose we have an experiment that generates m+1≥2 possible outcomes, {X0,...,Xm}, each occurring with non-negative probabilities {p0,...,pm} respectively. If sampling proceeded until n observations were made, then {X0,...,Xm} would have been multinomially distributed. However, if the experiment is stopped once X0 reaches the predetermined value x0 (assuming x0 is a positive integer), then the distribution of the m-tuple {X1,...,Xm} is negative multinomial. These variables are not multinomially distributed because their sum X1+...+Xm is not fixed, being a draw from a negative binomial distribution.

Properties

edit

Marginal distributions

edit

If m-dimensional x is partitioned as follows   and accordingly     and let  

The marginal distribution of   is  . That is the marginal distribution is also negative multinomial with the   removed and the remaining p's properly scaled so as to add to one.

The univariate marginal   is said to have a negative binomial distribution.

Conditional distributions

edit

The conditional distribution of   given   is  . That is,  

Independent sums

edit

If   and If   are independent, then  . Similarly and conversely, it is easy to see from the characteristic function that the negative multinomial is infinitely divisible.

Aggregation

edit

If   then, if the random variables with subscripts i and j are dropped from the vector and replaced by their sum,  

This aggregation property may be used to derive the marginal distribution of   mentioned above.

Correlation matrix

edit

The entries of the correlation matrix are    

Parameter estimation

edit

Method of Moments

edit

If we let the mean vector of the negative multinomial be   and covariance matrix   then it is easy to show through properties of determinants that  . From this, it can be shown that   and  

Substituting sample moments yields the method of moments estimates   and  

edit

References

edit
  1. ^ Le Gall, F. The modes of a negative multinomial distribution, Statistics & Probability Letters, Volume 76, Issue 6, 15 March 2006, Pages 619-624, ISSN 0167-7152, 10.1016/j.spl.2005.09.009.

Waller LA and Zelterman D. (1997). Log-linear modeling with the negative multi- nomial distribution. Biometrics 53: 971–82.

Further reading

edit

Johnson, Norman L.; Kotz, Samuel; Balakrishnan, N. (1997). "Chapter 36: Negative Multinomial and Other Multinomial-Related Distributions". Discrete Multivariate Distributions. Wiley. ISBN 978-0-471-12844-1.