To use the ARAR algorithm to predict 24 values of the accidental deaths data, open the file…
To use the ARAR algorithm to predict 24 values of the accidental deaths data, open the file DEATHS.TSM and proceed as described above. Selecting Minimize WN variance[max lag=26] gives the graph of the data and predictors shown in Figure 9.1. Rightclicking on the graph and then Info, we find that the selected memory shortening filter is (1 − .9779B^{12}). The fitted subset auto regression and the coefficients ξ_{j}of the overall whitening filter ξ(B) are shown below:
Optimal lags 
1 
3 
12 
13 

Optimal coeffs 
.5915 
.3822 
.3022 
.2970 

WN Variance: 
.12314E+06 




COEFFICIENTS OF OVERALL WHITENING FILTER: 

1.0000 
.5915 
.0000 
.2093 
.0000 

.0000 
.0000 
.0000 
.0000 
.0000 

.0000 
.0000 
.6757 
.2814 
.0000 

.2047 
.0000 
.0000 
.0000 
.0000 

.0000 
.0000 
.0000 
.0000 
.2955 

.2904 



