> # Poisson regression
> mdata  <-read.table("poisson.csv", header=T, sep=",")
> id <-mdata[,1]
> y  <-mdata[,2]
> x  <-mdata[,3]
> mydata <-data.frame(y,x)
> mypoi <-glm(y~x, family="poisson", data=mydata)
> summary(mypoi)
Call:
glm(formula = y ~ x, family = "poisson")

Deviance Residuals: 
     Min        1Q    Median        3Q       Max  
-11.2714   -2.7181   -0.8283    2.4182   11.3187  

Coefficients:
            Estimate Std. Error z value Pr(>|z|)    
(Intercept)  1.52733    0.06088   25.09   <2e-16 ***
x            0.97707    0.01615   60.51   <2e-16 ***
---
Signif. codes:  0 �e***�f 0.001 �e**�f 0.01 �e*�f 0.05 �e.�f 0.1 �e �f 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 5862.6  on 44  degrees of freedom
Residual deviance: 1215.6  on 43  degrees of freedom
AIC: 1469.4

Number of Fisher Scoring iterations: 5

>#**********************************************************************
># Negative Binomial regression
>#**********************************************************************
> library(MASS)
> mynegb <-glm.nb(y~x, data=mydata)
> summary(mynegb)

Call:
glm.nb(formula = y ~ x, data = mydata, init.theta = 2.1899289, 
    link = log)

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-2.5652  -0.8778  -0.3171   0.3069   1.6715  

Coefficients:
            Estimate Std. Error z value Pr(>|z|)    
(Intercept)  2.02627    0.23173   8.744   <2e-16 ***
x            0.81801    0.08334   9.815   <2e-16 ***
---
Signif. codes:  0 �e***�f 0.001 �e**�f 0.01 �e*�f 0.05 �e.�f 0.1 �e �f 1

(Dispersion parameter for Negative Binomial(2.1899) family taken to be 1)

    Null deviance: 153.975  on 44  degrees of freedom
Residual deviance:  48.945  on 43  degrees of freedom
AIC: 444.98

Number of Fisher Scoring iterations: 1


              Theta:  2.190 
          Std. Err.:  0.480 

 2 x log-likelihood:  -438.977 
>