AIC公式
[urlhttp://www.pinggu.org/bbs/viewth ... p;page=1#pid4216379][/url]

好的
呵呵不错谢谢

您好！我不知道如何在上面的代码上添加这公式，您帮我添进去吧！谢谢！

x <- c(rep(0,4),rep(1,3),rep(0,2),
rep(1,3),rep(0,2),rep(1,3),0,1,0)
n <- c(4,3,4,6,10,14,12,2,3,4,5,10,4,
5,9,5,10,4,7,4)
nx <- data.frame(n,x)
nx.glm <- glm(n ~ x, family="poisson", data=nx)
summary(nx.glm)
Call:
glm(formula = n ~ x, family = "poisson", data = nx)

Deviance Residuals:
     Min        1Q    Median        3Q       Max  
-1.75393  -0.49739   0.05042   0.46525   1.68649  

Coefficients:
            Estimate Std. Error z value Pr(>|z|)   
(Intercept)   1.3610     0.1601   8.499  < 2e-16 ***
x             0.7908     0.1931   4.096  4.2e-05 ***
---
Signif. codes:  0 ｀***＇ 0.001 ｀**＇ 0.01 ｀*＇ 0.05 ｀.＇ 0.1 ｀ ＇ 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 32.616  on 19  degrees of freedom
Residual deviance: 14.502  on 18  degrees of freedom
AIC: 89.951

Number of Fisher Scoring iterations: 4

##自行计算AIC
AIC = -2*sum(dpois(nx.glm$y, nx.glm$fitted.values, log=TRUE)*nx.glm$prior.weights)+2*nx.glm$rank
AIC   # 89.95149