Introduction

Time series can be found everywhere; if you analyze the stock market, sunspot occurrences, or river flows, you are observing phenomena that are stretched in time. It is almost inevitable that any data scientist throughout his or her career will deal with time series data at some point. In this chapter, we will see various techniques of handling, analyzing, and building models for time series.

The datasets for this chapter come from the web archive of river flows, which can be accessed here:

http://ftp.uni-bayreuth.de/math/statlib/datasets/riverflow

The archive is essentially a shell script that we processed to create the datasets for this chapter. In order to create the raw files from the archive, you can use Cygwin (on Windows) or Terminal on Mac/Linux and execute the following command (assuming that you save the archive in riverflows.webarchive):

sh riverflow.webarchive