Hi everyone,
I am trying to run a panel threshold regression to check the effect of financial inclusion on Economic growth.
I have used PCA to construct and index, and run the regression for a sample of developed and developing countries.
. xthreg GDPpcg FII inf pop to, qx(FII) rx(fd)
Estimating the threshold parameters: 1st ...... Done
Threshold estimator (level = 95):
-----------------------------------------------------
model | Threshold Lower Upper
-----------+-----------------------------------------
Th-1 | 0.8114 0.7757 0.9768
-----------------------------------------------------
Fixed-effects (within) regression Number of obs = 106
Group variable: country_id Number of groups = 53
R-squared: Obs per group:
Within = 0.6199 min = 2
Between = 0.2762 avg = 2.0
Overall = 0.1945 max = 2
F(6, 47) = 12.77
corr(u_i, Xb) = -0.9807 Prob > F = 0.0000
------------------------------------------------------------------------------
GDPpcg | Coefficient Std. err. t P>|t| [95% conf. interval]
-------------+----------------------------------------------------------------
FII | 1.795848 .5170819 3.47 0.001 .7556137 2.836083
inf | -.0262133 .1141041 -0.23 0.819 -.2557612 .2033346
pop | -2.115352 .5620811 -3.76 0.000 -3.246113 -.9845907
to | .2148181 .0648698 3.31 0.002 .0843169 .3453193
|
_cat#c.fd |
0 | .0119328 .0220283 0.54 0.591 -.0323825 .0562481
1 | -.0189764 .0246487 -0.77 0.445 -.0685633 .0306104
|
_cons | -13.0959 6.316101 -2.07 0.044 -25.80226 -.3895464
-------------+----------------------------------------------------------------
sigma_u | 13.138957
sigma_e | 2.1043823
rho | .97498921 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(52, 47) = 2.56 Prob > F = 0.0007
I am however having two problems:
1. When i try to split the sample to run the regression on the developed and developing countries separately its throwing an error.
. keep if classification == "developed"
(38 observations deleted)
. br
. xthreg GDPpcg FII inf pop to, qx(FII) rx(fd)
thestm(): 3301 subscript invalid
<istmt>: - function returned error
r(3301);
2. I am unable to use the financial inclusion Index as the threshold AND regime variable
. xthreg GDPpcg FII inf pop to, qx(FII) rx(FII)
Estimating the threshold parameters: 1st ...... thest(): 3200 conformability error
thestm(): - function returned error
<istmt>: - function returned error
r(3200);
Can someone please explain how i can fix this. It is essential for me to do a comparative analysis between the groups.
Thank you
I am trying to run a panel threshold regression to check the effect of financial inclusion on Economic growth.
I have used PCA to construct and index, and run the regression for a sample of developed and developing countries.
. xthreg GDPpcg FII inf pop to, qx(FII) rx(fd)
Estimating the threshold parameters: 1st ...... Done
Threshold estimator (level = 95):
-----------------------------------------------------
model | Threshold Lower Upper
-----------+-----------------------------------------
Th-1 | 0.8114 0.7757 0.9768
-----------------------------------------------------
Fixed-effects (within) regression Number of obs = 106
Group variable: country_id Number of groups = 53
R-squared: Obs per group:
Within = 0.6199 min = 2
Between = 0.2762 avg = 2.0
Overall = 0.1945 max = 2
F(6, 47) = 12.77
corr(u_i, Xb) = -0.9807 Prob > F = 0.0000
------------------------------------------------------------------------------
GDPpcg | Coefficient Std. err. t P>|t| [95% conf. interval]
-------------+----------------------------------------------------------------
FII | 1.795848 .5170819 3.47 0.001 .7556137 2.836083
inf | -.0262133 .1141041 -0.23 0.819 -.2557612 .2033346
pop | -2.115352 .5620811 -3.76 0.000 -3.246113 -.9845907
to | .2148181 .0648698 3.31 0.002 .0843169 .3453193
|
_cat#c.fd |
0 | .0119328 .0220283 0.54 0.591 -.0323825 .0562481
1 | -.0189764 .0246487 -0.77 0.445 -.0685633 .0306104
|
_cons | -13.0959 6.316101 -2.07 0.044 -25.80226 -.3895464
-------------+----------------------------------------------------------------
sigma_u | 13.138957
sigma_e | 2.1043823
rho | .97498921 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(52, 47) = 2.56 Prob > F = 0.0007
I am however having two problems:
1. When i try to split the sample to run the regression on the developed and developing countries separately its throwing an error.
. keep if classification == "developed"
(38 observations deleted)
. br
. xthreg GDPpcg FII inf pop to, qx(FII) rx(fd)
thestm(): 3301 subscript invalid
<istmt>: - function returned error
r(3301);
2. I am unable to use the financial inclusion Index as the threshold AND regime variable
. xthreg GDPpcg FII inf pop to, qx(FII) rx(FII)
Estimating the threshold parameters: 1st ...... thest(): 3200 conformability error
thestm(): - function returned error
<istmt>: - function returned error
r(3200);
Can someone please explain how i can fix this. It is essential for me to do a comparative analysis between the groups.
Thank you
