Median and its variants form the core measures of EDA and you would have got a hang of it by the first section. The visualization techniques of EDA also compose more than just the stem-and-leaf plot, letter values, and bagplot. As EDA is basically about your attitude and approach, it is important to realize that you can (and should) use any method that is instinctive and appropriate for the data on hand. We have also built our first regression model in the resistant line and seen how robust it is to the outliers. Smoothing data and median polish are also advanced EDA techniques that the reader is acquainted with from their respective sections.
EDA is exploratory in nature and its findings may need further statistical validations. The next chapter on statistical inference addresses what Tukey calls, confirmatory analysis. Especially, we look at techniques that give good point estimates of the unknown parameters. This is then backed with further techniques such as goodness-of-fit and...