I am running a panel regression (xtreg) on an unbalanced dataset with large N, small T. I want to check for specification errors. However, ovtest does not work after xtreg. In a previous post I read pooled OLS and ovtest should give you some idea about what is going on. On top of that, I performed RAMSEY manually. However, both test return opposite results. Did I perform RAMSEY well?
Code:
. regress pctcarbonintensitychange L.firmsize L.profitability L.leverage age L.capitalintensity
> L.CAPEX L.KZindex elektricitygenerator Carbonleakage i.twodigitsNACE i.Province_n SME public
> firm, cluster(Companynumber)
Linear regression Number of obs = 864
F(39, 182) = .
Prob > F = .
R-squared = 0.0720
Root MSE = .15726
(Std. Err. adjusted for 183 clusters in Companynumber)
----------------------------------------------------------------------------------------------
| Robust
pctcarbonintensitychange | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-----------------------------+----------------------------------------------------------------
firmsize |
L1. | -.0090182 .0065504 -1.38 0.170 -.0219427 .0039063
|
profitability |
L1. | -.0682747 .1258495 -0.54 0.588 -.3165864 .1800369
|
leverage |
L1. | .02627 .0228704 1.15 0.252 -.0188552 .0713953
|
age | -.000399 .0002034 -1.96 0.051 -.0008002 2.21e-06
|
capitalintensity |
L1. | .0162414 .0341511 0.48 0.635 -.0511416 .0836244
|
CAPEX |
L1. | -.0316673 .08159 -0.39 0.698 -.1926512 .1293166
|
KZindex |
L1. | -.0015618 .0038121 -0.41 0.683 -.0090833 .0059598
|
elektricitygenerator | -.2961985 .0393391 -7.53 0.000 -.3738179 -.2185792
Carbonleakage | .0243106 .0116731 2.08 0.039 .0012786 .0473426
|
twodigitsNACE |
10 | -.0195584 .0330305 -0.59 0.554 -.0847304 .0456136
11 | -.0577665 .0456884 -1.26 0.208 -.1479135 .0323806
13 | -.078182 .0420238 -1.86 0.064 -.1610986 .0047345
16 | -.1037839 .0400834 -2.59 0.010 -.1828718 -.0246961
17 | -.0351672 .0328799 -1.07 0.286 -.100042 .0297075
19 | -.0186398 .0417693 -0.45 0.656 -.1010541 .0637746
20 | -.0350844 .0309991 -1.13 0.259 -.0962482 .0260794
21 | -.1494211 .04283 -3.49 0.001 -.2339283 -.064914
22 | -.020242 .0390535 -0.52 0.605 -.0972978 .0568137
23 | -.0624852 .0302421 -2.07 0.040 -.1221555 -.002815
24 | -.0539425 .0324851 -1.66 0.099 -.1180383 .0101532
25 | -.0627991 .0349913 -1.79 0.074 -.13184 .0062417
27 | -.1412946 .0338315 -4.18 0.000 -.208047 -.0745422
28 | -.1198783 .0768268 -1.56 0.120 -.2714641 .0317074
29 | -.0832753 .044148 -1.89 0.061 -.1703829 .0038323
30 | -.1360702 .0325089 -4.19 0.000 -.2002129 -.0719275
35 | .290227 .0391526 7.41 0.000 .2129757 .3674783
42 | -.0630286 .0332746 -1.89 0.060 -.1286821 .0026249
46 | -.0535625 .0369998 -1.45 0.149 -.1265662 .0194412
47 | .014183 .0329183 0.43 0.667 -.0507676 .0791335
49 | -.06336 .0465584 -1.36 0.175 -.1552237 .0285036
52 | -.0684168 .0473292 -1.45 0.150 -.1618013 .0249676
61 | -.1777199 .0414566 -4.29 0.000 -.2595173 -.0959225
63 | -.0113177 .063323 -0.18 0.858 -.1362593 .1136238
70 | -.0499445 .0486463 -1.03 0.306 -.1459276 .0460387
72 | -.1255995 .034986 -3.59 0.000 -.1946297 -.0565693
81 | -.1359017 .035425 -3.84 0.000 -.2057981 -.0660052
|
Province_n |
Brabant Wallon / Waals Br.. | .0077831 .0270338 0.29 0.774 -.0455568 .0611231
Brussels | .0382777 .0206957 1.85 0.066 -.0025567 .0791121
East-Flanders | -.020138 .0153932 -1.31 0.192 -.0505101 .0102341
Hainaut / Henegouwen | .015691 .0170014 0.92 0.357 -.0178541 .0492362
Limburg / Limbourg | -.014696 .0157812 -0.93 0.353 -.0458336 .0164416
Liège / Luik | .0067641 .0196872 0.34 0.732 -.0320804 .0456087
Luxembourg | .0031876 .0168744 0.19 0.850 -.030107 .0364821
Namur / Namen | -.0135726 .0329515 -0.41 0.681 -.0785887 .0514435
Vlaams Brabant / Brabant .. | -.025935 .0234185 -1.11 0.270 -.0721417 .0202717
West-Flanders | .0209583 .0202837 1.03 0.303 -.0190631 .0609796
|
SME | -.034733 .0157155 -2.21 0.028 -.0657409 -.003725
publicfirm | .0194497 .0342806 0.57 0.571 -.0481888 .0870882
_cons | .2054266 .133863 1.53 0.127 -.0586962 .4695495
----------------------------------------------------------------------------------------------
. ovtest
Ramsey RESET test using powers of the fitted values of pctcarbonintensitychange
Ho: model has no omitted variables
F(3, 812) = 0.67
Prob > F = 0.5678
.
. *ramsey manually
. predict yhat1 if e(sample), xb
(183 missing values generated)
. generate yhat2 = yhat1^2
(183 missing values generated)
. generate yhat3 = yhat1^3
(183 missing values generated)
. generate yhat4 = yhat1^4
(183 missing values generated)
. xtreg pctcarbonintensitychange L.firmsize L.profitability L.leverage age L.capitalintensity L
> .CAPEX L.KZindex elektricitygenerator Carbonleakage i.twodigitsNACE i.Province_n SME publicfi
> rm yhat2 yhat3 yhat4, fe cluster(Companynumber)
Fixed-effects (within) regression Number of obs = 864
Group variable: Companynum~r Number of groups = 183
R-sq: Obs per group:
within = 0.0166 min = 1
between = 0.0017 avg = 4.7
overall = 0.0023 max = 5
F(9,182) = 2.56
corr(u_i, Xb) = -0.8775 Prob > F = 0.0086
(Std. Err. adjusted for 183 clusters in Companynumber)
----------------------------------------------------------------------------------------------
| Robust
pctcarbonintensitychange | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-----------------------------+----------------------------------------------------------------
firmsize |
L1. | .0159994 .0334022 0.48 0.633 -.049906 .0819048
|
profitability |
L1. | -.0839207 .1518765 -0.55 0.581 -.3835857 .2157444
|
leverage |
L1. | -.13617 .0827586 -1.65 0.102 -.2994597 .0271197
|
age | 0 (omitted)
|
capitalintensity |
L1. | .0807531 .1373122 0.59 0.557 -.1901755 .3516817
|
CAPEX |
L1. | -.0228027 .1008687 -0.23 0.821 -.2218251 .1762196
|
KZindex |
L1. | .0015868 .0051326 0.31 0.758 -.0085402 .0117138
yhat2 | -57.13183 34.73003 -1.65 0.102 -125.6571 11.39344
yhat3 | -90.23658 138.0355 -0.65 0.514 -362.5922 182.1191
yhat4 | 698.3039 458.6084 1.52 0.130 -206.569 1603.177
_cons | -.2031752 .6312734 -0.32 0.748 -1.448731 1.04238
-----------------------------+----------------------------------------------------------------
sigma_u | .19049121
sigma_e | .1601652
rho | .58584152 (fraction of variance due to u_i)
----------------------------------------------------------------------------------------------
. test yhat2 yhat3 yhat4
( 1) yhat2 = 0
( 2) yhat3 = 0
( 3) yhat4 = 0
F( 3, 182) = 4.04
Prob > F = 0.0083Timea De Wispelaere
0 Response to specification error in panel data
Post a Comment