Table AV.

Estimated STR models for secondary education (Regional analysis)

CASTILLA Y LEÓNCATALUÑACOM. VALENCIANAPAÍS VASCO
Linear modelSTR modelLinear modelSTR modelLinear modelSTR modelLinear modelSTR model
Transition variable: ENRStTransition variable: ENRStTransition variable: ENRSt-1Transition variable: ENRSt
GDPt-10.43 (0.10)  1.07 (0.28)0.38 (0.09)−0.46 (0.10)0.50 (0.14)−0.63 (0.16)
GDPt-2        
ENRSt 0.40 (0.15) 5.86 (1.72)   1.24 (0.36)
ENRSt-1   −0.51 (0.28) 0.74 (0.16)  
ENRSt-2  0.18 (0.06)  0.16 (0.05)0.08 (0.04) 
PHYt 0.10 (0.02)0.08 (0.03) 0.12 (0.03)0.10 (0.02)  
PHYt-1 −0.09 (0.01)0.07 (0.03)0.39 (0.10)    
PHYt-2     0.10 (0.02)  
LABSt        
LABSt-1     0.11 (0.03)  
LABSt-2   −0.36 (0.17)    
EXPt0.16 (0.03)0.46 (0.06)0.17 (0.03) 0.17 (0.03)0.59 (0.06)0.19 (0.04)0.80 (0.08)
EXPt-1   −1.23 (0.41)  −0.11 (0.05) 
EXPt-2 −0.25 (0.04)   −0.43 (0.07) −0.31 (0.07)
F(st)×GDPt-1   −1.20 (0.40) 0.47 (0.18) 0.63 (0.16)
F(st)×GDPt-2 2.14 (0.35)   −0.77 (0.22) −1.57 (0.38)
F(st)×ENRSt −0.44 (0.16) −5.66 (1.70) −1.73 (0.26) −1.80 (0.54)
F(st)×ENRSt-1       −1.27 (0.42)
F(st)×ENRSt-2     0.16 (0.05)  
F(st)×PHYt −0.39 (0.07) 0.06 (0.02) −0.21 (0.04)  
F(st)×PHYt-1 −0.09 (0.01) −0.37 (0.10)   0.15 (0.04)
F(st)×PHYt-2        
F(st)×LABSt 0.72 (0.14)      
F(st)×LABSt-1 −0.57 (0.13)   0.11 (0.03)  
F(st)×LABSt-2   0.48 (0.20) 0.25 (0.06) 0.95 (0.22)
F(st)×EXPt −0.46 (0.06) 0.22 (0.07)    
F(st)×EXPt-1 0.20 (0.07) 1.42 (0.44)    
F(st)×EXPt-2 −0.25 (0.04)   0.51 (0.14) 0.63 (0.18)
Γ 5.37 (2.01) 4.60 (2.21) 8.99 (3.64) 12.88 (5.02)
C 0.06 (0.00) 0.03 (0.01) 0.06 (0.00) 0.05 (0.00)
R¯2 0.890.630.870.690.940.460.85
s2/s2L 0.12 0.16 0.08 0.13
AUTO 3.00 (0.10) 5.87 (0.03) 3.24 (0.10) 2.47 (0.14)
NL 0.23 (0.93) 0.28 (0.96) 0.33 (0.89) 1.85 (0.40)
PC 0.87 (0.64) 1.78 (0.54) 0.62 (0.74) 2.87 (0.21)

Notes: All variables are first differences of the logarithmic transformation. Values after regression coefficients are SEs of the estimates; R-2 is the adjusted coefficient of determination; s2/s2L is the variance ratio of the residuals from the STR model and the linear regression; AUTO is the test for residual autocorrelation of order 1; NL is the test for no remaining nonlinearity; PC is the general parameter constancy test. Numbers in parentheses after misspecification tests statistics are p-values. This table reports the estimated models for secondary and tertiary education using the explanatory variables listed in the first column. For each educational level, the first column presents the linear model and the second column is for the STR specification. Regarding the nonlinear model, the upper part of the table corresponds to the “lower” regime (F = 0), and the next part introduces the interactions with the transition function [equation (1)]. The smoothness of the transition between regimes is given by the slope parameter γ and c represents the estimated threshold

Source: Own elaboration

or Create an Account

Close Modal
Close Modal