Model with two or more explanatory variables
The simple linear model with one explanatory variable can be extended with more terms.
Body fat
We have shown how extra variables can be added to a least squares equation to help predict the response.
However we have not investigated whether adding more variables actually improves predictions.
This type of question cannot be addressed until we introduce methodology to use data to ask questions about the process from which the data arose — a collection of methods called inference.