This is the author speaking here. What a great chapter! Yeah, I know I should not say that since I am the author of the book, nevertheless, I think the one you just completed was a relevant step towards your discovery of R for data mining. You are now able to:

• Fit a linear model in R, both having a single explanatory variable and multiple explanatory variables (univariate and multivariate) through the lm() function and assess its estimates through the summary() function
• Evaluate whether the linear regression model assumptions are met, through the durbinWatsonTest() and NCVtest() functions
• Perform principal component regression on your data through the pcr() function
• Perform stepwise regression through the stepAIC() function and evaluate its output
• Compare and interpret the output and performance of different regression models and evaluate whether your model is a reasonable way to describe the observed phenomenon

It's now time to take a closer look at what a model performance, introducing or examining in greater depth concepts such as R-squared, squared error, and the confusion matrix.