Hi,
I was testing the cross sectional independence test using Pesaran CD for the panel data of 51 states across US and time frame of 2000-2010. After runnin the fixed effect with time dummies (effects), I got a negative value for the test with the p value >1 (which is not possible).
I checked the paper on this http://www.stata-journal.com/article...article=st0113 but they didnt really talk about any negative value. I would appreciate it if someone could help me out here.
I am currently using the latest XTCSD version of 1.1.1.
. xtreg lgsr ue real_gdp income health_ins age1 age2 age4 male divorce alcohol i.year, fe
Fixed-effects (within) regression Number of obs = 561
Group variable: state Number of groups = 51
R-sq: within = 0.3508 Obs per group: min = 11
between = 0.0026 avg = 11.0
overall = 0.0013 max = 11
F(20,490) = 13.24
corr(u_i, Xb) = -0.3021 Prob > F = 0.0000
------------------------------------------------------------------------------
lgsr | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
ue | .016019 .0045147 3.55 0.000 .0071484 .0248895
real_gdp | .0003206 .0016599 0.19 0.847 -.0029409 .003582
income | .0021002 .0021859 0.96 0.337 -.0021948 .0063952
health_ins | -.0016709 .0015455 -1.08 0.280 -.0047076 .0013658
age1 | .0019928 .0038545 0.52 0.605 -.0055805 .0095661
age2 | .0023042 .0020005 1.15 0.250 -.0016263 .0062348
age4 | .0016173 .0012709 1.27 0.204 -.0008798 .0041145
male | .0066157 .0062386 1.06 0.289 -.0056421 .0188735
divorce | -.00114 .0042629 -0.27 0.789 -.0095157 .0072358
alcohol | .1834608 .0477275 3.84 0.000 .0896849 .2772367
|
year |
2001 | .0399855 .0147932 2.70 0.007 .0109195 .0690515
2002 | .0394005 .0196149 2.01 0.045 .0008608 .0779402
2003 | .0185931 .0195112 0.95 0.341 -.0197429 .0569291
2004 | .0319237 .0190439 1.68 0.094 -.005494 .0693415
2005 | .0318436 .0198853 1.60 0.110 -.0072274 .0709146
2006 | .0394558 .0200144 1.97 0.049 .0001311 .0787804
2007 | .0628746 .0216417 2.91 0.004 .0203526 .1053965
2008 | .086305 .0248275 3.48 0.001 .0375234 .1350865
2009 | .0538239 .0378625 1.42 0.156 -.020569 .1282169
2010 | .0990387 .0359585 2.75 0.006 .0283869 .1696905
|
_cons | 1.453003 .3710163 3.92 0.000 .7240242 2.181983
-------------+----------------------------------------------------------------
sigma_u | .30586385
sigma_e | .07074507
rho | .94921887 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(50, 490) = 80.69 Prob > F = 0.0000
. xtcsd, pesaran abs
Pesaran's test of cross sectional independence = -1.308, Pr = 0.1908
Average absolute value of the off-diagonal elements = 0.263
thank you !
I was testing the cross sectional independence test using Pesaran CD for the panel data of 51 states across US and time frame of 2000-2010. After runnin the fixed effect with time dummies (effects), I got a negative value for the test with the p value >1 (which is not possible).
I checked the paper on this http://www.stata-journal.com/article...article=st0113 but they didnt really talk about any negative value. I would appreciate it if someone could help me out here.
I am currently using the latest XTCSD version of 1.1.1.
. xtreg lgsr ue real_gdp income health_ins age1 age2 age4 male divorce alcohol i.year, fe
Fixed-effects (within) regression Number of obs = 561
Group variable: state Number of groups = 51
R-sq: within = 0.3508 Obs per group: min = 11
between = 0.0026 avg = 11.0
overall = 0.0013 max = 11
F(20,490) = 13.24
corr(u_i, Xb) = -0.3021 Prob > F = 0.0000
------------------------------------------------------------------------------
lgsr | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
ue | .016019 .0045147 3.55 0.000 .0071484 .0248895
real_gdp | .0003206 .0016599 0.19 0.847 -.0029409 .003582
income | .0021002 .0021859 0.96 0.337 -.0021948 .0063952
health_ins | -.0016709 .0015455 -1.08 0.280 -.0047076 .0013658
age1 | .0019928 .0038545 0.52 0.605 -.0055805 .0095661
age2 | .0023042 .0020005 1.15 0.250 -.0016263 .0062348
age4 | .0016173 .0012709 1.27 0.204 -.0008798 .0041145
male | .0066157 .0062386 1.06 0.289 -.0056421 .0188735
divorce | -.00114 .0042629 -0.27 0.789 -.0095157 .0072358
alcohol | .1834608 .0477275 3.84 0.000 .0896849 .2772367
|
year |
2001 | .0399855 .0147932 2.70 0.007 .0109195 .0690515
2002 | .0394005 .0196149 2.01 0.045 .0008608 .0779402
2003 | .0185931 .0195112 0.95 0.341 -.0197429 .0569291
2004 | .0319237 .0190439 1.68 0.094 -.005494 .0693415
2005 | .0318436 .0198853 1.60 0.110 -.0072274 .0709146
2006 | .0394558 .0200144 1.97 0.049 .0001311 .0787804
2007 | .0628746 .0216417 2.91 0.004 .0203526 .1053965
2008 | .086305 .0248275 3.48 0.001 .0375234 .1350865
2009 | .0538239 .0378625 1.42 0.156 -.020569 .1282169
2010 | .0990387 .0359585 2.75 0.006 .0283869 .1696905
|
_cons | 1.453003 .3710163 3.92 0.000 .7240242 2.181983
-------------+----------------------------------------------------------------
sigma_u | .30586385
sigma_e | .07074507
rho | .94921887 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(50, 490) = 80.69 Prob > F = 0.0000
. xtcsd, pesaran abs
Pesaran's test of cross sectional independence = -1.308, Pr = 0.1908
Average absolute value of the off-diagonal elements = 0.263
thank you !
Comment