- 16 Videos
- 56 Exercises
- 4 hours
- 4,245 Participants
- 4550 XP

**Instructor(s):**

Jeffrey Ryan is creator of xts and quantmod, as well as a multitude of other packages for R and finance. He is an original organizer of the annual R in Finance conference in Chicago, and currently works in the hedge fund space in Chicago.

Lore Dirick

Josiah Parry

Time series are all around us, from server logs to high frequency financial data. Managing and manipulating ordered observations is central to all time series analysis. The xts and zoo packages provide a set of powerful tools to make this task fast and mistake free. In this course, you will learn everything from the basics of xts to advanced tips and tricks for working with time series data in R.

xts and zoo are just two of the many different types of objects that exist in R. This chapter will introduce the basic objects in xts and zoo and their components, and offers examples of how to construct and examine the data.

- Welcome to the course! 50 xp
- Introducing the xts and zoo objects 50 xp
- What is an xts object? 50 xp
- More than a matrix 100 xp
- Your first xts object 100 xp
- Deconstructing xts 100 xp
- Time based indices 100 xp
- Importing, exporting and converting time series 50 xp
- Converting xts objects 100 xp
- Importing data 100 xp
- Exporting xts objects 100 xp

Now that you can create basic xts objects, it's time to see how powerful they can be. This chapter will cover the basics of one of the most useful features of xts: time based subsetting. From there you'll explore additional ways to extract data using time phrases, and conclude with how to do basic operations like adding and subtracting of your xts objects.

One of the most important parts of working with time series data involves creating derived time series. To do this effectively, it is critical to keep track of dates and times. In this chapter will look at how xts handles merging new columns and rows into existing data, how to deal with the inevitable missing observations in time series, and how to shift your series in time.

Now the fun begins! A very common usage pattern for time series is to calculate values for disjoint periods of time or aggregate values from a higher frequency to a lower frequency. For most series, you'll often want to see the weekly mean of a price or measurement. You may even find yourself looking at data that has different frequencies and you need to normalize to the lowest frequency. This chapter is where it all happens. Hang tight, and lets get going!

Now that you are comfortable with most of the core features, its time to explore some of the lesser known (but powerful!) aspects of working with xts. In this final chapter you will use the internals of the index to find repeating itervals, see how xts provides intuitive time zone support, and experiment with ways to explore your data by time - including identifying frequency and coverage in time. Let's finish this course!