|
|
Line 132: |
Line 132: |
|
|
|
|
|
If we imagine an urn containing <math>x_0</math> white balls (success) and <math>m</math> colors of balls numbering <math>(\alpha_1, \alpha_2, \ldots, \alpha_m) </math> respectively (also non-negative integers in this case) and we successively draw from the urn until we observe <math>x_0</math> white balls - and each time we draw from the urn and observe the color of the drawn ball we return two balls of the observed color back to the urn (in a [[Pólya urn model|Pólya urn scheme]]), then the resulting distribution of the number of observed m failure colors is Dirichlet negative multinomial. In the case where all parameters are positive integers, the distribution can clearly be written in purely in terms of [[factorials]] as opposed to [[gamma function]]s. |
|
If we imagine an urn containing <math>x_0</math> white balls (success) and <math>m</math> colors of balls numbering <math>(\alpha_1, \alpha_2, \ldots, \alpha_m) </math> respectively (also non-negative integers in this case) and we successively draw from the urn until we observe <math>x_0</math> white balls - and each time we draw from the urn and observe the color of the drawn ball we return two balls of the observed color back to the urn (in a [[Pólya urn model|Pólya urn scheme]]), then the resulting distribution of the number of observed m failure colors is Dirichlet negative multinomial. In the case where all parameters are positive integers, the distribution can clearly be written in purely in terms of [[factorials]] as opposed to [[gamma function]]s. |
|
|
|
|
===Dirichlet negative multinomial as a model for consumer behavior=== |
|
|
One application of the DNM distribution is for observed consumer transactions of m distinct brands with no special groupings. The model considered by Goodhardt, Ehrenberg and Chatfield<ref>Goodhardt, G.J., Ehrenberg, A.S.C. and Chatfield, C.(1984). The Dirichlet: a comprehensive model of buying |
|
|
behaviour. Journal of the Royal Statistical Society. Series A (General) 147, 621–655.</ref> assumes |
|
|
|
|
|
# Purchasing of a product class takes the form of a [[Poisson process]] for each consumer |
|
|
# There is heterogeneity in purchase rates among the consumers following a [[gamma distribution]] |
|
|
# The choice of m brands for each consumer follows a [[multinomial distribution]] |
|
|
# There is heterogeneity of the brand choice among the consumers follows a [[Dirichlet distribution]] |
|
|
|
|
|
The model has been found to account for a number of empirical generalisations, including [[Double jeopardy (marketing)|double jeopardy]], the duplication of purchase law, and natural monopoly. It has been shown to hold over different product categories, countries, time, and for both subscription and repertoire repeat-purchase markets. It has been described as one of the most famous empirical generalisations in marketing, |
|
|
|
|
|
|
==See also== |
|
==See also== |
Probability multivariate distribution
Notation |
|
---|
Parameters |
|
---|
Support |
|
---|
PDF |
where Γ(x) is the Gamma function. |
---|
Mean |
for |
---|
Variance |
for |
---|
MGF |
undefined |
---|
In probability theory and statistics, the Dirichlet negative multinomial distribution is a multivariate distribution on the non-negative integers. It is a multivariate extension of the beta negative binomial distribution. It is also a generalization of the negative multinomial distribution (NM(k, p)) allowing for heterogeneity or overdispersion to the probability vector. It is used in quantitative marketing research to flexibly model the number of household transactions across several brands.
If parameters of the Dirichlet distribution are , and if
where
then the marginal distribution of X is a Dirichlet negative multinomial distribution:
In the above, is the negative multinomial distribution and is the Dirichlet distribution.
Motivation
Dirichlet negative multinomial as a compound distribution
The Dirichlet distribution is a conjugate distribution to the negative multinomial distribution. This fact leads to an analytically tractable compound distribution.
For a random vector of category counts , distributed according to a negative multinomial distribution, the compound distribution is obtained by integrating on the distribution for p which can be thought of as a random vector following a Dirichlet distribution:
which results in the following formula:
where and are the dimensional vectors created by appending the scalars and to the dimensional vectors and respectively. We can write this equation explicitly as
Alternative formulations exist. One convenient representation[1] is
where and .
Properties
Marginal distributions
To obtain the marginal distribution over a subset of Dirichlet negative multinomial random variables, one only needs to drop the irrelevant 's (the variables that one wants to marginalize out) from the vector. The joint distribution of the remaining random variates is where is the vector with the removed 's.
Conditional distributions
If m-dimensional x is partitioned as follows
and accordingly
then the conditional distribution of on is where
and
- .
Conditional on the sum
The conditional distribution of a Dirichlet negative multinomial distribution on is Dirichlet-multinomial distribution with parameters and . That is
- .
Notice that the equation does not depend on or .
Correlation matrix
For the entries of the correlation matrix are
Heavy tailed
The Dirichlet negative multinomial is a heavy tailed distribution. It does not have a finite mean for and it has infinite covariance matrix for . It therefore has undefined moment generating function.
Aggregation
If
then, if the random variables with positive subscripts i and j are dropped from the vector and replaced by their sum,
Applications
Dirichlet negative multinomial as a urn model
The Dirichlet negative multinomial can also be motivated by an urn model in the case when is a positive integer. Consider a sequence of independent and identically distributed multinomial trials, each of which has outcomes. Call one of the outcomes a “success”, and suppose it has probability . The other outcomes – called "failures" - have probabilities . If the vector counts the m types of failures before the success is observed, then the have negative mulitnomial distribution with parameters .
If the parameters are themselves sampled from a Dirichlet distribution with parameters , then the resulting distribution of is Dirichlet negative multinomial. The resultant distribution has parameters.
If we imagine an urn containing white balls (success) and colors of balls numbering respectively (also non-negative integers in this case) and we successively draw from the urn until we observe white balls - and each time we draw from the urn and observe the color of the drawn ball we return two balls of the observed color back to the urn (in a Pólya urn scheme), then the resulting distribution of the number of observed m failure colors is Dirichlet negative multinomial. In the case where all parameters are positive integers, the distribution can clearly be written in purely in terms of factorials as opposed to gamma functions.
See also
References
- ^ Farewell, Daniel & Farewell, Vernon. (2012). Dirichlet negative multinomial regression for overdispersed correlated count data. Biostatistics (Oxford, England). 14. 10.1093/biostatistics/kxs050.