Determinants of Electricity Consumption Intensity in China: Analysis of Cities at Subprovince and Prefecture Levels in 2009

Table 4

Estimation results and variable structure for the finite mixture model^a.

Variable	Structure	Coefficient estimates
		Component 1	Component 2	Component 3	Component 4
		Predictor
Intercept	Varied	5461.407^***; (919.115)	379.656^***; (87.482)	1774.662^***; (158.055)	982.128^***; (61.198)
𝑋1	Varied	113.915^***; (40.749)	23.923^***; (2.821)	59.830^***; (6.378)	40.738^***; (3.021)
𝑋2	Varied	66.189; (92.219)	22.124^***; (6.895)	48.875^***; (12.326)	32.844^***; (4.232)
𝑋3	Fixed	−9.677^**; (4.896)	−9.677^**; (4.896)	−9.677^**; (4.896)	−9.677^**; (4.896)
𝑋4	Nested	19.305^***; (7.366)	19.305^***; (7.366)	17.365^**; (6.991)	17.365^**; (6.991)
𝑋5	Varied	−7.509^***; (1.315)	−0.287; (0.185)	−2.910^***; (0.414)	−0.759^***; (0.130)
𝑋6	Varied	−52.716; (35.108)	6.815^**; (2.750)	25.498^***; (7.160)	−7.531^***; (1.774)
Concomitant variable
Intercept	Varied	—	2.187^***; (0.441)	1.976^***; (0.473)	2.285^***; (0.516)
𝑋7	Varied	—	−0.032^***; (0.012)	−0.062^**; (0.024)	−0.186^**; (0.092)

^aThe standard errors of coefficient estimates are in parentheses. ^** and ^*** denote significance at 5% and 1% levels, respectively. All the explanatory variables are standardized.