Table 9.

Panel regression test results of MAS and MAS² on ROA and ROE.

Panel A: ROA (fixed effects)	Coefficient	t-statistics	p-value
Constant	0.8653	9.0428	0.0000**
MAS	0.3970	1.5214	0.1291
MAS2	−0.0646	−1.1222	0.2626
F-statistic = 30.6310 Prob. (F-statistic) = 0.0000**			Adjusted R2 = 0.4073 DW = 1.7776
Panel B: ROA (fixed effects)
Constant	22.7587	2.9473	0.0034**
MAS	0.2485	0.9603	0.3376
MAS2	−0.0478	−0.8549	0.3932
SIZE	−1.4645	−2.8356	0.0049**
ΔSALES	0.0226	4.7194	0.0000**
F-statistic = 29.6190 Prob. (F-statistic) = 0.0000**			Adjusted R2 = 0.4563 DW = 1.7750
Panel C: ROE (fixed effects)
Constant	0.9670	7.2160	0.0000**
MAS	0.6209	1.7169	0.0869
MAS2	−0.0990	−1.2418	0.2152
F-statistic = 29.5367 Prob. (F-statistic) = 0.0000**			Adjusted R2 = 0.3982 DW = 1.7771
Panel D: ROE (fixed effects)
Constant	37.7472	3.5480	0.0004**
MAS	0.3702	1.0386	0.2998
MAS2	−0.0708	−0.9189	0.3588
SIZE	−2.4611	−3.4585	0.0006**
ΔSALES	0.0313	4.7497	0.0000**
F-statistic = 29.4089 Prob. (F-statistic) = 0.0000**			Adjusted R2 = 0.4545 DW = 1.7783

Note: DW is the Durbin–Watson (1950) statistic, $DW={\sum }_{t=2}^N{({\hat{e}}_t-{\hat{e}}_{t-1})}^2/{\sum }_{t=1}^N{\hat{e}}_t^2$ and ${\hat{e}}_t$ are the least squares residuals from the regression equation under consideration. Based on the Durbin–Watson bounds test, there is no residual autocorrelation if DW is greater than the upper critical value bound, which approximately equals 1.78 at the 5% significance level.

^**Represent the significance level of 1%.