Session Ready
Exercise

The autocorrelation function

Autocorrelations can be estimated at many lags to better assess how a time series relates to its past. We are typically most interested in how a series relates to its most recent past.

The acf(..., lag.max = ..., plot = FALSE) function will estimate all autocorrelations from 0, 1, 2,..., up to the value specified by the argument lag.max. In the previous exercise, you focused on the lag-1 autocorrelation by setting the lag.max argument to 1.

In this exercise, you'll explore some further applications of the acf() command. Once again, the time series x has been preloaded for you and is shown in the plot on the right.

Instructions
100 XP
  • Use acf() to view the autocorrelations of series x from 0 to 10. Set the lag.max argument to 10 and keep the plot argument as FALSE.
  • Copy and paste the autocorrelation estimate (ACF) at lag-10.
  • Copy and paste the autocorrelation estimate (ACF) at lag-5.
    Loading