# varpt: variance for the Poisson-Tweedie family of distributions In poistweedie: Poisson-Tweedie exponential family models

## Description

Variance for the Poisson-Tweedie family of distributions

## Usage

 `1` ```varpt(mu, p, theta0) ```

## Arguments

 `p` is a real index related to a precise model. `mu` the mean. `theta0` the canonical parameter.

## Details

variance for the Poisson-Tweedie family of distributions

## Author(s)

Cactha David Pechel, Laure Pauline Fotso and Celestin C Kokonendji Maintainer: Cactha David Pechel ( <[email protected]>)

`dpoistweedie`, `ppoistweedie`
 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24``` ```## plot of variance layout(matrix(1:1,1,1)) layout.show(1) mu <- seq(0.001,6,l=100) var <-varpt(mu,p=5000,theta0=-150) plot(mu, var, type = "l", col = "green", lwd=1,main="variance(p,mu,theta0=-150)") grid(nx=1,ny=1, lty=1,lwd=2) lines(mu,varpt(mu,p=1,theta0=-150), type = "l", col = "blue", lwd=1) lines(mu,varpt(mu,p=2,theta0=-150), type = "l", col = "black", lwd=1) lines(mu,varpt(mu,p=1.5,theta0=-150), type = "l", col = "yellow", lwd=1) lines(mu,varpt(mu,p=2.5,theta0=-150), type = "l", col = "cyan", lwd=1) lines(mu,varpt(mu,p=3,theta0=-150), type = "l", col = "magenta", lwd=1) segments(4,2.5,4.5,2.5,col="blue" ) text(5,2.5,"p=1",cex=0.8) segments(4,2,4.5,2,col="yellow" ) text(5,2,"1.5",cex=0.8) segments(4,1.5,4.5,1.5,col= "black") text(5, 1.5,"p=2",cex=0.8) segments(4,1,4.5,1,col="cyan" ) text(5, 1,"p=2.5" ,cex=0.8) segments(4,0.5,4.5,0.5,col="magenta" ) text(5, 0.5,"p=3" ,cex=0.8) segments(4,0,4.5,0,col= "green" ) text(5, 0,"p=5000",cex=0.8) ```