21 Going above and beyond

 

This chapter covers

  • Consolidating your learning
  • Managing difficult forecasting problems
  • Exploring beyond time series forecasting
  • Sources of time series datasets

First of all, congratulations on making it to the end of this book! It has been quite a journey to get here, and it required a lot of your time, effort, and attention.

You have gained a lot of skills for time series forecasting, but there is, of course, a lot still to learn. The objective of this chapter is to summarize what you’ve learned and outline what else you can achieve with time series data. I’ll also encourage you to keep practicing your forecasting skills by listing various sources of time series data.

The real challenge lies ahead of you, as you apply your knowledge to problems, either at work or as a side project, where the solutions are unknown to you. It is important that you gain confidence in your skills, which can only come from experience and practicing often. It is my hope that this chapter will inspire you to do so.

21.1 Summarizing what you’ve learned

Our very first step in time series forecasting was to define a time series as a set of data points ordered in time. You also quickly learned that the order of the data must remain untouched for our forecasting models to make sense. This means that data measured on Monday must always come after Sunday and before Tuesday. Therefore, no shuffling of the data is allowed when splitting it into training and testing sets.

21.1.1 Statistical methods for forecasting

21.1.2 Deep learning methods for forecasting

21.1.3 Automating the forecasting process

21.2 What if forecasting does not work?

21.3 Other applications of time series data

21.4 Keep practicing