Time series analysis in R: part1

Every year we are expected to do a peer observation exercise, where we all have a colleague observe an aspect of our teaching, and provide feedback on it. You can have someone sit in one of your lectures, but you can also do something a bit more theoretical. Last year, for instance, I looked at the module website of one of my colleagues. It is actually quite useful; if you make one of those websites yourself, everything makes sense to you, but someone who looks at this fresh comes in with little expectation, and might find things puzzling.

This year I agreed to look at a computer practical one of my colleagues was providing. It is a practical in R. And I said that I was probably quite suitable to that; I am not a natural in R, so the assignment would have to be rather clear for me to be able to follow. And the practical was about time series analysis; that is such a useful thing to be able to do! I have done it in the past, with different software, and I thought it would be excellent if I would learn to do it in R. 

With the dissertation presentations out of the way, I sat down to make it happen. I downloaded his materials and started RStudio. Let the fun begin!

The practical came in two pieces. The first one was basically about familiarisation with wave functions and what alterations they can come with, and how you calculate and plot that in RStudio. The second was about actual data analysis. 

I didn’t mind that it starts simple. When it comes to R, I am self-taught, and also a sporadic user. Every time I use it I have to pretty much start from zero, so I'm quite happy to start simple here too. And I am very slow! But I was making progress. 

No rocket science yet, but something is happening

I started before Easter, but I continued in my home office. If it’s rubbish weather and your knee hurts you might as well do some work! Maybe you can then find some time for fun when knee and weather have improved a bit. And on Easter Monday I finished the first part. That felt like a triumph! And I don’t know how I will experience the second part. I suspect the science will be more serious, but will that result in much more complicated R scripts? Quite possibly not! But one way to find out…

End of part 1!