. 2015 Dec 28;36:154–171. doi: 10.1016/j.najef.2015.12.002

Table 3.

Estimation of stock return of semiconductor industry.

Model	PSTR-CDS	PSTR-VIX	PSTR-TED	Linear
Transition variable	CDS	VIX	TED spread	–
Parameter	r = 1, m = 2, d = 6	r = 1, m = 2, d = 5	r = m = 1, d = 6	–
γ	2.212	225.7	0.003	–
c₁	14.33	12.82	0.249	–
c₂	127.4	14.70	–	–

R_mt − R_ft
θ₁	1.446 (27.07)^***	1.829 (19.3)^***	1.067 (23.3)^***	0.733 (28.8)^***
$θ_{1^{'}}$	−0.09 (−1.442)	−0.48 (−4.87)^**	0.336 (5.78)^***	–

R_SMBt
θ₂	0.089 (1.349)	0.743 (9.45)^***	−0.078 (−1.34)	1.075 (33.0)^***
$θ_{2^{'}}$	0.253 (2.910)^**	−0.521 (−5.86)^***	0.443 (5.65)^***	–

R_HMLt
θ₃	0.142 (3.375)^**	0.243 (3.78)^**	0.190 (4.47)^***	0.113 (5.27)^***
$θ_{3^{'}}$	−0.308 (−5.52)^***	−0.367 (−5.25)^**	−0.294 (−5.71)^***	–

AIC	4.970	5.085	5.068	6.59
BIC	4.978	5.093	5.075	6.64
R-squared	–	–		0.263

Note: r and m denote the number of transition functions and location parameters, respectively. d is the lag length of transition variable. The digit in parenthesis is t-statistic. The three optimal models are chosen by employing minimum AIC and BIC.

*Significance at 10% level.

^**

Significance at 5% level.

^***

Significance at 1% level.