R Tutorial: Importing, exporting and converting time series

Want to learn more? Take the full course at https://learn.datacamp.com/courses/manipulating-time-series-data-with-xts-and-zoo-in-r at your own pace. More than a video, you'll learn hands-on coding & quickly apply skills to your daily work. --- In the last video we looked at creating xts objects from scratch. These were rather contrived, simply to illustrate the mechanics of xts construction and what xts objects looked like internally. In most real life cases, you'll be working with data that already exists - usually from some other process. Maybe it is a time series from a colleague in a different data class. In other cases you may find yourself importing data from an external source which may meet all the criteria you need for xts but it is coming from a file instead of another R object. In this chapter, well look at converting types using xts, reading data into R as an xts object, as well as exporting xts objects from R for other uses. First let's take a look at the most useful and simple way to convert most objects you'll encounter in R into xts. This will work in 90% of cases, as xts was designed from the beginning to make working with R's myriad time series and time classes as easy and flexible as possible. To illustrate how easy this is, we'll use the famous sunspots dataset that ships with R. sunspots is a ts object, which is fairly challenging to work with as it is regular - i.e. fixed intervals, but is less intuitive in its structure. To convert, we only need to use as.xts and you see the series now is well structured and looks like you might expect a time series to look. To import data from outside, we can follow a similar pattern. Here, we can read data into R using built in functions such as read.table, and coerce into xts at that point. As we mentioned earlier, since xts is a proper subclass of zoo, we can also leverage the powerful tools zoo provides to make life even easier. read.zoo is a great tool to read in data as a time series, and o