Hi,
I have the follow situation in a ancova analysis:
y (count data)
x1 (continuous data)
x2 (qualitative data)
x2 have 10 levels (a,b,c,d,e,f,g,h,i,j)
I make the model
m1 <- glm(y~x1*x2, family=poisson)
then I want to remove the levels h, i and j (make these levels with
intercept = 0 and slope = 0)
A make a new datafile without these levels. This new table is smallest than
the original table.
I refit the model:
m2 <- glm(y~x1*x2, family=poisson)
Then I want to compare m1 with m2, I need to know if the removing of the 3
levels are significant. But howto compare this if the total GL and the total
deviance are different.
Existe another mean to make this? Maybe using offset? How?
Thanks for all
Ronaldo
--
You can observe a lot just by watching. -- Yogi Berra
--
| // | \\ [*****************************][*******************]
|| ( ? ? ) [Ronaldo Reis J?nior ][PentiumIII-600 ]
| V [UFV/DBA-Entomologia ][HD: 30 + 10 Gb ]
|| / \ [36571-000 Vi?osa - MG ][RAM: 128 Mb ]
| /(.''`.)\ [Fone: 31-3899-2532 ][Video: SiS620-8Mb ]
||/(: :' :)\ [chrysopa at insecta.ufv.br ][Modem: Pctel-onboar]
|/ (`. `'` ) \[ICQ#: 5692561 ][Kernel: 2.4.18 ]
|| ( `- ) [*****************************][*******************]
||| _/ \_Powered by GNU/Debian W/Sarge D+ || Lxuser#: 205366
