The regression models that we have looked at till now have always presumed a single independent variable and with "linear" coefficients. It is much more likely when investigating cause and effect relationships that there are perhaps many variables that contribute or perhaps using a non-linear relationship between the unknown coefficients. To tease you to consider taking another course that covers multi-variate regression, in this section we briefly consider a two-variable model. We also consider an interesting example that illustrates the danger in using models to estimate values well beyond the range of the relevant data that has been used to create the model.