In this chapter, we looked at the many and interesting aspects of time series analysis in Python with Pandas and statsmodels, how they handle the data, and some of the basic manipulation functions that are available. We also looked at the concept of stationarity, how to test your time series for it, and how to transform a non-stationary series into a stationary one. You also found out the various patterns and components that time series can be built up by, and finally, we went through how to create ARIMA models and predict future values based on previous historical data.

This chapter concludes the book. We have covered many different analysis techniques and general statistical knowledge and how to use them in Python to your benefit. With the knowledge in this book, you can start exploring data, any kind of data. In addition to these chapters, there is an appendix. In Appendix, More on Jupyter Notebook and matplotlib Styles, I will look at Jupyter Notebook tips and extensions (plugins). I will also provide some links to further resources, including various data repositories for you to find data to download and create your own hypothesis to test.

As mentioned before, it is important to take the time to play around with data and try different algorithms and compare the results. I hope that you have also realized that much of the work in data analysis happens before actually applying the analysis method/algorithm to the data and making a figure with the results, and that making good figures that show all the results without getting bloated is very difficult. In general, with the analysis of real-world data, the process is difficult, and when done, you have to present it all in an easy way for everyone else to understand. With the content taught in this book, you should be able to produce a solid analysis of almost any data and appealing figures in your report that clearly highlight your results.