Generalized-Linear-Models.md

@@ -27,14 +27,25 @@ When constructing a generalized linear model, three major decisions must be made
 1. **Specifying the Random Component:** 
     An appropriate probability distribution for the response variable must be chosen. This 
     should be any member from the *natural exponential family* distributions: 
+
     - Continuous variables: 
       * Normal 
       * Gamma 
       * Inverse Gaussian 
+
     - Discrete variables: 
       * Poisson 
       * Bernoulli 
       * Binomial 
-2. 
+
+2. ** Specifying the Systematic Component (Linear Predictor):**
+    A linear combination of predictor variables should be constructed as the linear 
+    predictor of the model. 
+
+3. ** Choosing a Link Function:** 
+    Choosing an appropriate function which maps the mean of the response variable onto the 
+    linear predictor.
+
+