Table 4.

Specification of the best ARMA model estimated for each NO${}_2$ and PM₁₀ time series and corresponding parameter estimates and p-values (see Eq. (2)).

	Sensor ID 6761 - AR(1) model
Est.	${\varphi }_1=$ 0.463
p-value	$<$0.001
	Sensor ID 6781 - MA(1) model
Est.	${\theta }_1=$ 0.673
p-value	$<$0.001
	Sensor ID 30163 - MA(3) model
Est.	${\theta }_1=$ 0.737	${\theta }_2=$ 0.656	${\theta }_3=$ 0.230
p-value	$<$0.001	$<$0.001	0.074
	Sensor ID 6951 - ARMA(2,1) model
Est.	${\varphi }_1=1.187$	${\varphi }_2=-0.665$	${\theta }_1=-0.321$
p-value	$<$0.001	$<$0.001	0.047
	Sensor ID 9961 - MA(3) model
Est.	${\theta }_1=$ 0.472	${\theta }_2=$ 0.087	${\theta }_3=-0.202$	${\theta }_4=-0.480$
p-value	$<$0.001	0.530	0.198	$<$0.001