Obtaining the Final Univariate Model

Based on the above discussion, different models were fit, including seasonal differencing of 3 and 4 years, differencing of 1 year, and model orders between 1 and 3 parameters for both the seasonal and nonseasonal components. After trying different models, the final model was determined to be 

The normal probability plot of the residuals and the autocorrelation of the residuals are shown in Fig. 5.21. Both results show that the residuals are normally distributed and white. In the normal probability plot, the tails deviate a bit from what would be desirable, but given that this is real data, such behaviour is inevitable. 

A comparison between the predicted and measured temperatures is shown in Fig. 5.22. As before, the model gets the overall trends correct, but the individual predictions are not very good. 

The following are references that provide additional information about the topic  

General Time Series Analysis these sources also often contain information about transfer function-based analysis  

---