| genpoisson {VGAM} | R Documentation |
Estimation of the two parameters of a generalized Poisson distribution.
genpoisson(llambda = "logit", ltheta = "loge",
elambda=list(), etheta=list(),
ilambda = 0.5, itheta = NULL, zero = NULL)
llambda |
Parameter link function for lambda.
See Links for more choices.
The parameter lies in the unit interval, therefore the
logit link is a reasonable default.
|
ltheta |
Parameter link function for theta.
See Links for more choices.
The parameter is positive, therefore the default is the log link.
|
elambda, etheta |
List. Extra argument for each of the links.
See earg in Links for general information.
|
ilambda |
Optional initial value for lambda. |
itheta |
Optional initial value for theta. |
zero |
An integer vector, containing the value 1 or 2. If so,
lambda or theta respectively are modelled as an intercept only.
The default value NULL means both linear/additive predictors
are modelled as functions of the explanatory variables.
|
See Consul (1989) for details.
An object of class "vglmff" (see vglmff-class).
The object is used by modelling functions such as vglm,
and vgam.
Convergence problems may occur when lambda is very close to 0.
This distribution is useful for dispersion modelling.
T. W. Yee
Jorgensen, B. (1997) The Theory of Dispersion Models. London: Chapman & Hall
Consul, P. C. (1989) Generalized Poisson Distributions: Properties and Applications, Marcel Dekker.
y = rpois(n=100, lam=24) fit = vglm(y ~ 1, genpoisson, trace=TRUE) fitted(fit)[1:5] mean(y) summary(fit) coef(fit, matrix=TRUE) Coef(fit)