Consider again the examination of wildeat strikes in the eoal industry
data.
( a) Examine the regression diagnostics for the Poisson regression
model on nonunionized loeations. Is there evidenee of problems
related to any of the individual mines?
(b) Fit a negative binomial regression model to the unionized 10-
eation data. Does this model have different implieations than
the quasi-likelihood model? Is the mine that appeared unusual
in the earlier analysis unusual when fitting a negative binomial
model?