Summary

In this chapter, we discussed several types of data such as cross-sectional, time series, and panel data. We delved into the special properties that make time series data special. Several examples and working code in Python have been discussed to give an understanding of how exploratory data analysis can be performed on time series to visualize its properties. We also described autocorrelation and partial autocorrelation and graphical techniques to detect these in a time series. The topics discussed in this chapter give us the stage for a more detailed discussion for working on time series data in Python.
In the next chapter, you will learn how to read more complex data types in time series and use such information for more in-depth exploratory data analysis. We will revisit autocorrelation in the context of stationarity of time series. Statistical methods to detect autocorrelation would be discussed. We would also discuss importance of stationarity and describe different differencing and averaging methods for stationarizing a non-stationary time series. Additive and multiplicative models of time decomposition for estimating trend and seasonality are discussed.