Question 1.
The dataset gives information about the number of military coups in sub-Saharan Africa and various political and geographical information. Use the following R commands to understand the variables in this dataset:
> install.packages("faraway")
> library(faraway)
> help(africa)
Develop a simple,
but well-fitting, model for the number of coups. Give an interpretation of the
effect of the variables you include in your model on the response.
Question 2.
The
data was collected in a salmonella reverse mutagenicity assay. The predictor
is the dose level of quinoline and the response is the numbers of revertant
colonies of TA98 Salmonella observed on each of three replicate plates. Show
that a Poisson GLM is inadequate and that some overdispersion must be allowed
for. Do not forget to check out other reasons for a high deviance.
Question 3.
The
data was collected from subjects in Denmark. Build a Poisson model to
describe how marital status (i.e., single, married, or divorced) changes with
age, taking care to give an understandable interpretation of the important
parameter estimates. Under your model, predict the probability that a Dane, aged
55, is divorced.
¡@