next up previous
Next: Autocorrelation Up: Results Previous: Periodogram

Standard deviation of aggregates

Figure 8 indicates that a plot of against has a slope close to zero, implying , for . Here, is dependent upon N, the number of points in the time series used. From figure 8, it is seen that grows apparently without limit, as N increases, at least for N up to . This in turn implies that is self-similar over time scales up to at least token passes.

  
Figure 8: Log. of normalised standard deviation of aggregates plot for time series, for , and points respectively. Again, M = 100 and the load is 60%. The more points that are used in the calculation, the larger the range over which the slope of the graph is close to zero (implying ).



Jonathan Deane, and David Jefferies
Mon Jun 3 13:27:09 BST 1996