Lecture 13

Julia Schedler

Agenda

  • SARIMA simulation activity

  • FPP sections 8.1-8.3 (Also 9.10 if time)

  • Sign up for Github Copilot

SARIMA simulation activity setup

  • Log on to the lab computer

  • Launch R and install whichever package you want to use

  • Let me know when you are ready

SARIMA simulation activity

  • Choose a pecan
  • Whether \(p,q, P, Q\) are nonzero are determined on the pecan (dot goes on the left)
  • Choose \(d\) yourself
  • I will come around and assign you seasonal or nonseasonal
  • If you get seasonal, choose \(D\) yourself
  • simulate from your model using arima.sim, then try to recover your model using tools from class.

SARIMA simulation activity

  • Hide your simulation code and leave just the output you generated
  • Switch computers with a partner
  • Can you recover the model your partner used to simulate the data?

Record your result

  • In the big ledger book, use a stamp of your choice
  • Use purple if you were not able to recover the model order, and green if you were

FPP 8.1-8.3

We will work through these examples collaboratively.

Github copilot

  • Make a github account (or log in)– USE A PERSONAL EMAIL (or add one to your account)
  • Sign up for the copilot beta
  • In Rstudio, go to Tools –> Global Options –> copilot and sign in with your github account
  • When you get approved, you will be able to use copilot in Rstudio (you might have to go through the Global options setup again)

If time– Sunspot data

Code 
library(astsa)
tsplot(sunspots)
Code 
?sunspots

https://www.sidc.be/SILSO/faq2

python code