Imagine you measure not just one number on a randomly chosen person—say height—but also weight, blood pressure, and resting heart rate. These measurements tend to move together in a predictable way. To capture the full picture of how they vary together, we need a random vector (a bundle of several random variables) and a multivariate distribution that describes their joint behaviour. In this chapter we build the tools to work with many variables at once, from joint density functions to the important multivariate normal distribution, and finish with a useful technique called order statistics.