This notebook uses MLE to estimate the parameters of a GLM(Generalized Linear Model) with Poisson distribution.
The data used is 'bird_count.csv' which contains the number of birds counted in different years.
The model uses 'yr' as predictor variable and 'count' as response variable. By minimizing the negative log likelihood, we can get the parameters of the model.
Using the estimated parameters, we generate 'count' samples for each year 3 times and save them to 'samples.csv'.