Use Partial Autocorrelation to calculate and plot the correlation between observations in a time series. The partial autocorrelation is the correlation between observations in a time series that is not accounted for by all of the shorter intervals between those observations. For example, the partial autocorrelation for a lag of 6 is only the correlation that is not accounted for by lags 1 through 5. The plot of partial autocorrelations is called the partial autocorrelation function (PACF). View the PACF to guide your choice of terms to include in an ARIMA model.
For example, an employment analyst uses a partial autocorrelation analysis to help create a model to study the trends in employment in three industries across five years.
To perform a partial autocorrelation analysis, choose .