Dynamic Decomposition Analysis and Forecasting of Energy Consumption in Shanxi Province Based on VAR and GM (1, 1) Models

Table 2

Statistical results of ADF test.

Variable	Test type (c,t,p)	ADF statistics	10% critical values	5% critical values	P value	Result
Ln(EC)	(c,1,9)	0.8753	-3.3103	-3.7332	0.9993	Nonstationary
D(Ln(EC))	(c,0,0)	-3.9381	-2.6336	-2.9919	0.0063	Stationary
Ln(PGDP)	(c,1,1)	-2.1390	-3.2431	-3.6122	0.4997	Nonstationary
D(Ln(PGDP))	(c,0,0)	-3.3258	-2.6336	-2.9919	0.0297	Stationary
Ln(P)	(c,1,0)	-2.6685	-3.2381	-3.6032	0.2565	Nonstationary
D(Ln(P))	(c,0,0)	-5.0175	-2.6336	-2.9919	0.0005	Stationary
Ln(T)	(c,1,0)	-3.2312	-3.2381	-3.6032	0.1012	Nonstationary
D(Ln(T))	(c,0,1)	-5.0734	-2.6388	-2.9981	0.0005	Stationary
Ln(URB)	(c,1,0)	-2.8535	-3.2381	-3.6032	0.1930	Nonstationary
D(Ln(URB))	(c,0,0)	-6.9982	-2.6336	-2.9919	0.0001	Stationary

Note: (c,t,P) is the test type, where c is the intercept, t is the trend, p is the lag length, and Zero value means there is no such item. Lag order selected by the SIC criterion. , , and , respectively, denote the rejection of the null hypothesis at 1%, 5%, and 10% significance level.