*
* Example for Canadian Employment data pp 140-142
*
cal(q) 1961:1
open data caemp.dat
data(format=prn,org=columns) 1961:1 1994:4
*
graph(footer="Figure 6.9 Canadian Employment Index")
# caemp
*
* You'll note that the two standard error bands on the graph aren't flat like
* those shown in figure 6.10. The flat standard error bands are based upon the
* asymptotic distribution of the correlations assuming that the series is white
* noise. BJIDENT computes the distribution of the autocorrelation rho(k) assuming
* that the correlations for lags 1,...,k-1 are equal to the estimated values. For
* series with high correlations for the low lags, these expand rather quickly.
*
@bjident(number=12) caemp