SQL for Data Analysis

Chapter 3. Time Series Analysis

Now that I’ve covered SQL and databases and the key steps in preparing data for analysis, it’s time to turn to specific types of analysis that can be done with SQL. There are a seemingly unending number of data sets in the world, and correspondingly infinite ways in which they could be analyzed. In this and the following chapters, I have organized types of analysis into themes that I hope will be helpful as you build your analysis and SQL skills. Many of the techniques to be discussed build on those shown in Chapter 2 and then on the preceding chapters as the book progresses. Time series of data are so prevalent and so important that I’ll start the series of analysis themes here.

Time series analysis is one of the most common types of analysis done with SQL. A time series is a sequence of measurements or data points recorded in time order, often at regularly spaced intervals. There are many examples of time series data in daily life, such as the daily high temperature, the closing value of the S&P 500 stock index, or the number of daily steps recorded by your fitness tracker. Time series analysis is used in a wide variety of industries and disciplines, from statistics and engineering to weather forecasting and business planning. Time series analysis is a way to understand and quantify how things change over time.

Forecasting is a common goal of time series analysis. Since time only marches forward, future values can be expressed as a function of ...