Time Series Modeling - Water Quality Management: Assessment and Interpretation - page 63

Fig. 4.5 Diagram of time series for each of water quality parameters and their predictions (the
predictions are based on 1 month ahead projections)

4.17.1.2 Selected Models of Few Water Quality Parameters in Zir-e-Pol
Station

Ca ++

As described in Fig. 4.5 the proper model for calcium (Ca þþ ) was ARIMA.
(2,0,0)(0,0,0) The relation of the model is as follows:

Z t =2 : 186 þ 1 : 02Z t 1 0 : 338Z t 2 þ a t ð 4 : 45 Þ

where Z t is the amount of calcium, and a t stands for error. The standard error is
0.331 according to comparison approaches. Akaike Information Criteria (AIC) and
Schwartz Bayesian Information Criteria are less than the other models. Also cor-
relation coef cient is 0.95 which is proper.

p-Value is 0.99 which indicates that the model is excellent (Table 4.6 ).

SO 4 --

Figure 4.5 shows variations of sulfate parameter and the best model for SO 4 is an

ARIMA (2,0,0) (1,0,0) S with seasonal components (Table 4.4 ).