I've computed a loglinear model on a categorical dataset. I would like to
test whether an interaction can be dropped by comparing the log-likelihoods
from two models(the model with the interaction vs. the model without).
Since R does not immediately print the log-likelihood when I use the glm
I think you need to learn about deviances, which R does print.
Log-likelihoods are only defined up to additive constants. In this case
the conventional constant differs if you view this as a Poisson or as a
product-multinomial log-linear model, and R gives you the log-likelihood
for a Poisson
You're right. I do need to learn more. I never learned null/residual
deviance. I know the deviance is equivalent to an anova decompostion. But
I've never dealt with it seperated like this.
I understand deviance as the difference between two model's log-likelihood
difference between them
Alternatively generate the log-likelihood using the sum(dpois(y,
fitted(model), log = TRUE))
Regards
Ross Darnell
Doxastic wrote:
You're right. I do need to learn more. I never learned null/residual
deviance. I know the deviance is equivalent to an anova decompostion.
But I've
Thanks. I used this and it gave me the same result as the logLik function.
The reason I ask is the SAS output gives me a loglik = 1089. R gives me
-298.09583. Both for my reduced model. For the saturated (or complex)
model, SAS gives me an loglik = 1143. R gives me -298.1993. The problem
At 07:30 AM 5/2/2007, Doxastic wrote:
Thanks. I used this and it gave me the same result as the logLik function.
The reason I ask is the SAS output gives me a loglik = 1089. R gives me
-298.09583. Both for my reduced model. For the saturated (or complex)
model, SAS gives me an loglik = 1143.
