Table 2.

The Regression Model based on Equation (1).

Linear regression Number of obs = 368   F(21, 346) = 29.16   Prob > F = 0.0000   R-squared = 0.5959   Root MSE = 115.83   Robust
y	$\mathit{\beta } \mathbf{Coefficients}$	Standard Error	t	P > t	(95% Confidence Interval)
var1	0.0436	0.0080	5.50	0.000 ***	0.0280–0.0592
var2	0.0356	0.0203	1.75	0.081 *	−0.0044–0.0755
var3	0.0042	0.0008	5.39	0.000 ***	0.0027–0.0058
var4 = 1	105.3848	32.6609	3.23	0.001 ***	41.1459–169.6237
var5 =
2	−105.664	18.3378	−5.76	0.000 ***	−141.7317–−69.5964
3	−142.515	40.0669	−3.56	0.000 ***	−221.3204–−63.7096
var6 = 1	44.8862	16.5621	2.71	0.007 ***	12.3111–77.4613
var12 = 1	171.0338	59.9637	2.85	0.005 ***	53.0946–288.9731
var9 = 1	23.8974	32.1457	0.74	0.458	−39.3281–87.1230
var10 = 1	13.3541	38.2681	0.35	0.727	−61.9132–88.6215
var8 * var7
0 1	−94.1936
1 0	60.0584	34.6160	−2.72	0.007 ***	−162.278–−26.1093
1 1	0 (omitted)	38.9043	1.54	0.124	−16.4603–136.577
var11 * var7
0 1	−67.3168
1 0	159.3906	32.0474	−2.10	0.036 *	−130.3491–−4.2845
1 1	0 (omitted)	39.2444	4.06	0.000 ***	82.2029–236.5783
var13 * var7
1 1	257.8345	64.8942	3.97	0.000 ***	130.1978–385.4712
2 0	59.5823	59.4744	1.00	0.317	−57.3946–176.5593
2 1	283.7637	72.0074	3.94	0.000 ***	142.1364–425.391
3 0	−24.1865	66.7344	−0.36	0.717	−155.4427–107.0697
3 1	300.5646	96.5215	3.11	0.002 ***	110.7219–490.4072
4 0	−25.5099	65.3923	−0.39	0.697	−154.1263–103.1066
4 1	270.0733	80.1628	3.37	0.001 ***	112.4057–427.741
_cons	63.0924	77.3488	0.82	0.415	−89.0406–215.2255

Note: The subscripts *, **, and *** refer to significance levels for two-tailed tests at p < 0.1, p < 0.05, and p < 0.01, respectively. For dummy and categorical variables, we notice the confidence intervals are somewhat wide in the result, which is mainly caused by the small and imbalanced sample size (especially for categorical variables). This can only be improved with a larger sample size or the inclusion of other key variables once additional relevant data and information becomes available.