I am using a panel of 27 countries for the period 1995-2015. I have both fixed country and year effects.
I am confused with the -xtreg, fe- command. The Wooldridge test shows serial correlation among my variables hence I plan to use -vce (robust)-. When I use -xtreg, fe- command, I get the following results:
But when I use the command, -xtreg, fe vce(cluster)-, I get this result:
How should I know which of the two results are correct, given the Wooldridge tests show presence of correlation?
I am confused with the -xtreg, fe- command. The Wooldridge test shows serial correlation among my variables hence I plan to use -vce (robust)-. When I use -xtreg, fe- command, I get the following results:
Code:
xtreg PATR1 l.PATR1 preelection right preelectionright left preelectionleft
> pit corporate inflation gdp unemployment govtexp cashtransfers wage waged ur
> banisation dependency dependencyold dependencyyoung popbelow14 pop1564 pop65
> andabove i.year, fe
note: dependencyyoung omitted because of collinearity
note: pop65andabove omitted because of collinearity
Fixed-effects (within) regression Number of obs = 432
Group variable: country Number of groups = 27
R-sq: Obs per group:
within = 0.6760 min = 7
between = 0.9627 avg = 16.0
overall = 0.9319 max = 19
F(38,367) = 20.15
corr(u_i, Xb) = 0.7308 Prob > F = 0.0000
------------------------------------------------------------------------------
PATR1 | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
PATR1 |
L1. | .7558233 .0350376 21.57 0.000 .6869236 .824723
|
preelection | .0091026 .0040812 2.23 0.026 .001077 .0171282
right | .0053177 .0028711 1.85 0.065 -.0003281 .0109635
preelecti~ht | -.0104984 .0052195 -2.01 0.045 -.0207623 -.0002345
left | .0082178 .0030789 2.67 0.008 .0021633 .0142723
preelecti~ft | -.0141748 .005544 -2.56 0.011 -.0250769 -.0032728
pit | .1268696 .1099153 1.15 0.249 -.0892732 .3430124
corporate | -.0105342 .0231331 -0.46 0.649 -.0560242 .0349558
inflation | .0259074 .0398219 0.65 0.516 -.0524004 .1042151
gdp | -.000022 .0003177 -0.07 0.945 -.0006468 .0006028
unemployment | .0177927 .0297763 0.60 0.551 -.0407608 .0763461
govtexp | -.0020769 .0311107 -0.07 0.947 -.0632546 .0591008
cashtransf~s | -.0551216 .2794898 -0.20 0.844 -.6047241 .4944808
wage | -.0008461 .0003622 -2.34 0.020 -.0015582 -.0001339
waged | .0077053 .0095505 0.81 0.420 -.0110752 .0264858
urbanisation | .0333709 .0597809 0.56 0.577 -.0841853 .150927
dependency | -.2216378 1.350354 -0.16 0.870 -2.87704 2.433765
dependency~d | -.3829479 1.581253 -0.24 0.809 -3.492401 2.726506
dependency~g | 0 (omitted)
popbelow14 | -.0045714 .0237036 -0.19 0.847 -.0511833 .0420406
pop1564 | -.0111767 .0182918 -0.61 0.542 -.0471467 .0247933
pop65andab~e | 0 (omitted)
|
year |
1998 | -.0083945 .00528 -1.59 0.113 -.0187775 .0019884
1999 | -.0033408 .0054345 -0.61 0.539 -.0140274 .0073458
2000 | .0202285 .0052874 3.83 0.000 .0098311 .0306259
2001 | -.0020623 .00529 -0.39 0.697 -.0124649 .0083402
2002 | .0018505 .0053045 0.35 0.727 -.0085804 .0122815
2003 | -.0042204 .0055222 -0.76 0.445 -.0150795 .0066387
2004 | .0065276 .0056636 1.15 0.250 -.0046096 .0176648
2005 | .0062469 .0062668 1.00 0.320 -.0060764 .0185702
2006 | .0009627 .0062121 0.15 0.877 -.011253 .0131785
2007 | .0050787 .0064215 0.79 0.430 -.0075488 .0177063
2008 | .0049623 .0071692 0.69 0.489 -.0091355 .0190602
2009 | .0038424 .0079866 0.48 0.631 -.0118628 .0195477
2010 | .0063669 .0075091 0.85 0.397 -.0083994 .0211332
2011 | .0122426 .0078224 1.57 0.118 -.0031398 .027625
2012 | .0116071 .0079234 1.46 0.144 -.0039738 .027188
2013 | .0069815 .0081282 0.86 0.391 -.0090021 .0229651
2014 | .0091839 .0087559 1.05 0.295 -.0080341 .026402
2015 | .0110639 .0087771 1.26 0.208 -.0061959 .0283237
|
_cons | 1.053637 1.67542 0.63 0.530 -2.240992 4.348265
-------------+----------------------------------------------------------------
sigma_u | .02317097
sigma_e | .01308116
rho | .75831327 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(26, 367) = 1.87 Prob > F = 0.0069
Code:
xtreg PATR1 l.PATR1 preelection right preelectionright left preelectionleft
> pit corporate inflation gdp unemployment govtexp cashtransfers wage waged ur
> banisation dependency dependencyold dependencyyoung popbelow14 pop1564 pop65
> andabove i.year, fe cluster (country)
note: dependencyyoung omitted because of collinearity
note: pop65andabove omitted because of collinearity
Fixed-effects (within) regression Number of obs = 432
Group variable: country Number of groups = 27
R-sq: Obs per group:
within = 0.6760 min = 7
between = 0.9627 avg = 16.0
overall = 0.9319 max = 19
F(26,26) = .
corr(u_i, Xb) = 0.7308 Prob > F = .
(Std. Err. adjusted for 27 clusters in country)
------------------------------------------------------------------------------
| Robust
PATR1 | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
PATR1 |
L1. | .7558233 .0522036 14.48 0.000 .6485173 .8631294
|
preelection | .0091026 .0068769 1.32 0.197 -.005033 .0232382
right | .0053177 .0022296 2.39 0.025 .0007347 .0099007
preelecti~ht | -.0104984 .0074715 -1.41 0.172 -.0258563 .0048594
left | .0082178 .0033452 2.46 0.021 .0013417 .0150939
preelecti~ft | -.0141748 .0081703 -1.73 0.095 -.0309692 .0026195
pit | .1268696 .159619 0.79 0.434 -.2012319 .4549711
corporate | -.0105342 .03536 -0.30 0.768 -.0832177 .0621494
inflation | .0259074 .0793233 0.33 0.747 -.1371441 .1889588
gdp | -.000022 .0003674 -0.06 0.953 -.0007771 .0007332
unemployment | .0177927 .0306367 0.58 0.566 -.045182 .0807673
govtexp | -.0020769 .0307051 -0.07 0.947 -.0651921 .0610383
cashtransf~s | -.0551216 .2976797 -0.19 0.855 -.6670111 .5567678
wage | -.0008461 .0006323 -1.34 0.192 -.0021457 .0004536
waged | .0077053 .0071374 1.08 0.290 -.0069658 .0223764
urbanisation | .0333709 .1019212 0.33 0.746 -.1761311 .2428728
dependency | -.2216378 1.121984 -0.20 0.845 -2.527908 2.084633
dependency~d | -.3829479 1.300487 -0.29 0.771 -3.056137 2.290242
dependency~g | 0 (omitted)
popbelow14 | -.0045714 .0190782 -0.24 0.813 -.0437871 .0346444
pop1564 | -.0111767 .0122005 -0.92 0.368 -.0362552 .0139017
pop65andab~e | 0 (omitted)
|
year |
1998 | -.0083945 .0040291 -2.08 0.047 -.0166764 -.0001127
1999 | -.0033408 .0044405 -0.75 0.459 -.0124684 .0057868
2000 | .0202285 .0050044 4.04 0.000 .0099418 .0305152
2001 | -.0020623 .0028425 -0.73 0.475 -.0079052 .0037805
2002 | .0018505 .0033912 0.55 0.590 -.0051203 .0088213
2003 | -.0042204 .004646 -0.91 0.372 -.0137704 .0053296
2004 | .0065276 .0050992 1.28 0.212 -.003954 .0170091
2005 | .0062469 .0056749 1.10 0.281 -.0054181 .0179118
2006 | .0009627 .0035196 0.27 0.787 -.0062719 .0081973
2007 | .0050787 .0049606 1.02 0.315 -.005118 .0152755
2008 | .0049623 .0067603 0.73 0.469 -.0089337 .0188584
2009 | .0038424 .008422 0.46 0.652 -.0134692 .021154
2010 | .0063669 .0070507 0.90 0.375 -.0081261 .0208598
2011 | .0122426 .0078361 1.56 0.130 -.0038648 .02835
2012 | .0116071 .0084582 1.37 0.182 -.0057789 .0289931
2013 | .0069815 .0092998 0.75 0.460 -.0121347 .0260976
2014 | .0091839 .0105311 0.87 0.391 -.0124631 .030831
2015 | .0110639 .0109036 1.01 0.320 -.0113486 .0334765
|
_cons | 1.053637 1.096562 0.96 0.345 -1.200379 3.307653
-------------+----------------------------------------------------------------
sigma_u | .02317097
sigma_e | .01308116
rho | .75831327 (fraction of variance due to u_i)
------------------------------------------------------------------------------