Understanding the Generalized Linear Model algorithm

Generalized Linear Model (GLM), as its name suggests, is a flexible way of generalizing linear models. It was formulated by John Nelder and Robert Wedderburn as a way of combining various regression models into a single analysis with considerations given to different probability distributions. You can find their detailed paper (Nelder, J.A. and Wedderburn, R.W., 1972. Generalized linear models. Journal of the Royal Statistical Society: Series A (General), 135(3), pp.370-384.) at https://rss.onlinelibrary.wiley.com/doi/abs/10.2307/2344614.

Now, you may be wondering what linear models are. Why do we need to generalize them? What benefit does it provide? These are relevant questions indeed and they are pretty easy to understand without diving too deep into the mathematics. Once we break down the logic, you will notice that the concept of GLM is pretty easy to understand.

So, let’s start by understanding the basics...