Poisson-Gamma Distribution with GLM

Share:

Description

The function fits a mixed Poisson distribution, in which the random parameter follows Gamma distribution. As the method of estimation Expectation-maximization algorithm is used. In M-step the GLM is applied.

Usage

1
pg.dist.glm(variable, lambda.start, gamma.par.start, epsylon, n)

Arguments

variable

The count dependent variable in the regression.

lambda.start

The starting value of lambda parameter of Poisson distribution. Default to 1.

gamma.par.start

The starting value of delta parameter of Gamma distribution. Default to 1.

epsylon

Default to epsylon = 10^(-8)

n

The integer value for the Laguerre quadrature. Default to 100.

Details

This function provides estimated parameters of the model N|θ \sim Poisson(λ θ) where θ is a latent variable comes from Gamma distribution with one parameter γ. The pdf of Gamma is of the form f_θ(θ)=\frac{γ^γ}{Γ(γ)}θ^{γ-1}\exp(-γθ) . The parameter λ is determined by the intercept through log-link λ=\exp(β_0).

Value

lambda

fixed effect in mixed Poisson distribution

gamma.par

the parameter of mixing Gamma distribution

n.iter

n

likelihood.values

values of log-likelihood

References

Karlis, D. (2005). EM algorithm for mixed Poisson and other discrete distributions. Astin bulletin, 35(01), 3-24.

Examples

1
2
3
4
library(MASS)	
var = quine$Days[quine$Days<11]		
pGamma = pg.dist.glm(variable=var)	
print(pGamma)