Interpret the lag for Partial Autocorrelation

The lag is the number of time periods that separate the ordered data. Minitab uses the lags to calculate the partial autocorrelation coefficients. The maximum number of lags (as suggested by Box and Jenkins) is n/4 for a series with fewer than 240 observations or + 45 for a series with more than 240 observations, where n is the number of observations.