Generalised linear models and the analysis of count data
Not all data fits the assumptions of linear models. For instance, the errors may not be normally distributed and the terms may not combine additively to determine the dependent variable. In count data, the errors might be Poisson distributed and independent variables might combine multiplicatively to determine the count observed. Fortunately, an analysis similar to a linear model can be used in these cases.
Intellectual Climate Fund
- School of Life Sciences