Simulating ARMA models
As we saw in the video, any stationary time series can be written as a linear combination of white noise. In addition, any ARMA model has this form, so it is a good choice for modeling stationary time series.
R provides a simple function called arima.sim()
to generate data from an ARMA model. For example, the syntax for generating 100 observations from an MA(1) with parameter .9 is arima.sim(model = list(order = c(0, 0, 1), ma = .9 ), n = 100)
. You can also use order = c(0, 0, 0)
to generate white noise.
In this exercise, you will generate data from various ARMA models. For each command, generate 200 observations and plot the result.
This is a part of the course
“ARIMA Models in R”
Exercise instructions
- Use
arima.sim()
andplot()
to generate and plot white noise. - Use
arima.sim()
andplot()
to generate and plot an MA(1) with parameter .9. - Use
arima.sim()
andplot()
to generate and plot an AR(2) with parameters 1.5 and -.75.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Generate and plot white noise
WN <-
# Generate and plot an MA(1) with parameter .9
MA <-
# Generate and plot an AR(2) with parameters 1.5 and -.75
AR <-