Estimate the autocorrelation function (ACF) for an autoregression
What if you need to estimate the autocorrelation function from your data? To do so, you'll need the acf()
command, which estimates autocorrelation by exploring lags in your data. By default, this command generates a plot of the relationship between the current observation and lags extending backwards.
In this exercise, you'll use the acf()
command to estimate the autocorrelation function for three new simulated AR series (x
, y
, and z
). These objects have slope parameters 0.5, 0.9, and -0.75, respectively, and are shown in the adjoining figure.
This exercise is part of the course
Time Series Analysis in R
Exercise instructions
- Use three calls to
acf()
to calculate the ACF forx
,y
, andz
, respectively.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Calculate the ACF for x
acf(___)
# Calculate the ACF for y
# Calculate the ACF for z