|
|
|
front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |review |
 |
The data was
analyzed by Poisson log linear regression.
Explanatory and
response variables fall into categories (e.g., cases aged 40-49
years fall into the same age group).
The population
size is taken into account to calculate the incidence rates:
Number of cases observed, divided by the number of person-years under
observation (population).
The Poisson
distribution is used as an approximation of the Binomial
distribution for incidence rates, since the population size is much
larger than the number of cases.
This regression
belongs to the family of Generalized Linear Models (GLM): The
log. transformation is used to make the parameters‘ estimation
easier, because the model is linear on the logarithmic scale (see next
slide).
Poisson log
linear regressions and other GLMs are available in most softwares:
SAS, STATA, GLIM.
|