Variable Selection of High-Dimensional Spatial Autoregressive Panel Models with Fixed Effects

Table 5

Different models fitting China’s carbon emission dataset.

Coefficient	SARPF			PF
	SCAD	ALASSO	Nonpenalty	SCAD	ALASSO	Nonpenalty
	0.2080	0.2000	0.2065	0.0000	0.0000	0.0000
-PU	0.2385	0.1978	0.2231	0.0000	0.0000	0.0672
-EC	0.3476	0.3457	0.3434	0.3648	0.3674	0.3530
-EI	0.5908	0.5942	0.6006	0.6416	0.6397	0.6847
-SU	0.0000	0.0000	0.0243	0.0000	0.0000	0.0191
-GDP	0.1327	0.1514	0.1413	0.2428	0.2448	0.2275
-P	0.2222	0.2016	0.2354	0.1893	0.1798	0.2736
-IPC	0.0668	0.0659	0.0680	0.0810	0.0789	0.0835
-FIEISE	−0.0288	−0.0249	−0.0311	0.0000	0.0000	−0.0292
-VFT	0.0000	0.0000	0.0063	0.0000	0.0000	0.0235
-PITI	0.0000	0.0000	−0.0282	0.0000	0.0000	0.0269
R square	0.8410	0.8413	0.8413	0.8245	0.8244	0.8299

¹SARPF: SARP model with fixed effects and PF: panel model with fixed effects.