Table 4 Regression analyses Variables...

Table 4

Regression analyses

Variables	Model 1: OLS: OS	Model 2: Quantile models			Model 3: Probit: EARNDIS	Model 4: 2SLS: OS
Variables	Model 1: OLS: OS	Q_0.25	Q_0.50	Q_0.75	Model 3: Probit: EARNDIS	Model 4: 2SLS: OS
EARNDIS	7.319**	1.198	6.854**	13.591***	–	7.448**
EARNDIS	(3.834)	(1.997)	(3.463)	(7.807)	–	(4.130)
GFC	−16.693***	−5.113**	−12.300***	−9.743*	0.632	−16.679***
GFC	(4.553)	(2.544)	(3.615)	(4.989)	(0.836)	(3.527)
INSTOWN	10.619*	3.091	7.267	15.020*	0.040	−10.660**
INSTOWN	(10.823)	(3.067)	(5.451)	(7.807)	(1.235)	(3.526)
UR	−4.556	−1.618	−3.240	0.109	−0.363	−4.619*
UR	(2.806)	(1.455)	(3.471)	(4.192)	(1.269)	(2.400)
GROP	1.592	0.018	2.694	−0.177	−0.398	1.555
GROP	(4.024)	(2.926)	(4.378)	(6.184)	(1.269)	(4.762)
PUBOWN	−50.235***	−4.794	−25.077***	−65.864***	5.143**	−50.362***
PUBOWN	(10.827)	(5.224)	(8.933)	(15.546)	(2.330)	(8.429)
OFFSES	−15.894***	−5.632**	−12.272**	−24.315***	−0.389	−15.963***
OFFSES	(5.093)	(2.650)	(5.860)	(9.165)	(0.960)	(5.158)
LNOFFSZ	−9.089***	−1.180	−5.991***	−10.657***	0.959***	−9.072***
LNOFFSZ	(2.135)	(1.020)	(2.271)	(3.062)	(0.370)	(1.847)
DBOARD	−5.752	−2.587	−6.713*	−15.890**	3.438***	−5.728
DBOARD	(4.014)	(2.226)	(3.963)	(7.541)	(1.079)	(3.712)
DLY	−0.465**	−0.279***	−0.483***	−0.479**	−0.016	−0.466***
DLY	(0.203)	(0.075)	(0.157)	(0.226)	(0.024)	(0.149)
MKTRET	−0.628	−0.046	−0.300	−0.717	−0.059	−0.623**
MKTRET	(0.423)	(0.222)	(0.335)	(0.579)	(0.048)	(0.314)
LNAGE	–	–	–	–	0.119**	–
LNAGE	–	–	–	–	(0.258)	–
LOCKUP	–	–	–	–	−8.140***	–
LOCKUP	–	–	–	–	(1.267)	–
C	215.220***	41.181	148.876***	266.284***	−16.318***	214.549***
C	(36.743)	(17.475)	(38.845)	(52.342)	(6.010)	(31.290)
Observation	466	466	466	466	466	466
Adjusted R²/Pseudo R²	0.224	0.055	0.096	0.209	0.679	0.222
F-statistics	13.209***					12.897***
Durbin–Watson	1.275					1.278
VIF Range	1.093–2.283

Variables	Model 1: OLS: OS	Model 2: Quantile models			Model 3: Probit: EARNDIS	Model 4: 2SLS: OS
Variables	Model 1: OLS: OS	Q_0.25	Q_0.50	Q_0.75	Model 3: Probit: EARNDIS	Model 4: 2SLS: OS
EARNDIS	7.319**	1.198	6.854**	13.591***	–	7.448**
EARNDIS	(3.834)	(1.997)	(3.463)	(7.807)	–	(4.130)
GFC	−16.693***	−5.113**	−12.300***	−9.743*	0.632	−16.679***
GFC	(4.553)	(2.544)	(3.615)	(4.989)	(0.836)	(3.527)
INSTOWN	10.619*	3.091	7.267	15.020*	0.040	−10.660**
INSTOWN	(10.823)	(3.067)	(5.451)	(7.807)	(1.235)	(3.526)
UR	−4.556	−1.618	−3.240	0.109	−0.363	−4.619*
UR	(2.806)	(1.455)	(3.471)	(4.192)	(1.269)	(2.400)
GROP	1.592	0.018	2.694	−0.177	−0.398	1.555
GROP	(4.024)	(2.926)	(4.378)	(6.184)	(1.269)	(4.762)
PUBOWN	−50.235***	−4.794	−25.077***	−65.864***	5.143**	−50.362***
PUBOWN	(10.827)	(5.224)	(8.933)	(15.546)	(2.330)	(8.429)
OFFSES	−15.894***	−5.632**	−12.272**	−24.315***	−0.389	−15.963***
OFFSES	(5.093)	(2.650)	(5.860)	(9.165)	(0.960)	(5.158)
LNOFFSZ	−9.089***	−1.180	−5.991***	−10.657***	0.959***	−9.072***
LNOFFSZ	(2.135)	(1.020)	(2.271)	(3.062)	(0.370)	(1.847)
DBOARD	−5.752	−2.587	−6.713*	−15.890**	3.438***	−5.728
DBOARD	(4.014)	(2.226)	(3.963)	(7.541)	(1.079)	(3.712)
DLY	−0.465**	−0.279***	−0.483***	−0.479**	−0.016	−0.466***
DLY	(0.203)	(0.075)	(0.157)	(0.226)	(0.024)	(0.149)
MKTRET	−0.628	−0.046	−0.300	−0.717	−0.059	−0.623**
MKTRET	(0.423)	(0.222)	(0.335)	(0.579)	(0.048)	(0.314)
LNAGE	–	–	–	–	0.119**	–
LNAGE	–	–	–	–	(0.258)	–
LOCKUP	–	–	–	–	−8.140***	–
LOCKUP	–	–	–	–	(1.267)	–
C	215.220***	41.181	148.876***	266.284***	−16.318***	214.549***
C	(36.743)	(17.475)	(38.845)	(52.342)	(6.010)	(31.290)
Observation	466	466	466	466	466	466
Adjusted R²/Pseudo R²	0.224	0.055	0.096	0.209	0.679	0.222
F-statistics	13.209***					12.897***
Durbin–Watson	1.275					1.278
VIF Range	1.093–2.283

Note(s): LNAGE denotes the natural logarithm of a firm's age. LOCKUP denotes the dummy variable of the lock-up period whereby the IPOs issued under the 180-day lock-up regime (effective from the year 2009) are represented by “1”, whereas “0” is for the IPOs issued under the 360-day lock-up regime. The asterisks ***, ** and * signify significance at 1%, 5% and 10%, respectively. The quantile regression estimation uses the alternative Epanechnikov kernel function and Hall–Sheather's bandwidth

Source(s): Authors' own work

[ViewLarge]