Hey all, I am attempting to replicate my results achieved in another program within R (so I can expand my options for methods). I am trying to run a GLM (Family = Poisson) for count data in R. Some of my variables are factors and I am under the impression that the function glm() cannot run a model with Poisson dist and factors?? Here is why I think this, if I run two models using glm(), one treating a variable (e.g., Harvest.Factor) as a Factor and the other treating a variable as continuous (e.g., Harvest.Cont) I get the same results in R. Is there a package that will allow me to run a GLM using a Poisson dist and factor variables? Thanks for any help! -Chris -- View this message in context: http://r.789695.n4.nabble.com/GLM-Help-tp3949104p3949104.html Sent from the R help mailing list archive at Nabble.com.
GLM Help
2 messages · CES, Weidong Gu
Are you sure your variables are categorical or numeric? Of course, glm differentiates these two kinds of variables. For example, I ran the same variable with different modes, the results are very different.
dat<-data.frame(y=rpois(100,5),xf=as.factor(sample(1:4,100,replace=T))) glm(y~xf,data=dat,family=poisson)
Call: glm(formula = y ~ xf, family = poisson, data = dat)
Coefficients:
(Intercept) xf2 xf3 xf4
1.60944 -0.12783 -0.11878 -0.09746
...
glm(y~as.numeric(xf),data=dat,family=poisson)
Call: glm(formula = y ~ as.numeric(xf), family = poisson, data = dat)
Coefficients:
(Intercept) as.numeric(xf)
1.59047 -0.02673
Weidong Gu
On Fri, Oct 28, 2011 at 4:21 PM, CES <smit4155 at umn.edu> wrote:
Hey all, ?I am attempting to replicate my results achieved in another program within R (so I can expand my options for methods). I am trying to run a GLM (Family = Poisson) for count data in R. Some of my variables are factors and I am under the impression that the function glm() cannot run a model with Poisson dist and factors?? Here is why I think this, if I run two models using glm(), one treating a variable (e.g., Harvest.Factor) as a Factor and the other treating a variable as continuous (e.g., Harvest.Cont) I get the same results in R. Is there a package that will allow me to run a GLM using a Poisson dist and factor variables? Thanks for any help! -Chris -- View this message in context: http://r.789695.n4.nabble.com/GLM-Help-tp3949104p3949104.html Sent from the R help mailing list archive at Nabble.com.
______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.