I was checking my model for heteroscedasticity, and weird enough, I found a different outcome for the white test and the breusch pagan test. In my opinion, his is rather uncommon. Can anyone clarify what happened here and how I should interpret this ambiguous result?
Code:
. regress pctchangecarbonintensity firmsize profitability leverage age capitalintensity CAPEX K
> Zindex elektricitygenerator Carbonleakage industry10 industry11 industry13 industry16 industr
> y17 industry19 industry20 industry21 industry22 industry23 industry24 industry25 industry28 i
> ndustry29 industry30 industry35 industry42 industry46 industry47 industry49 industry52 indust
> ry63 industry70 industry72 industry81 WestFlanders Hainaut Antwerp Brussels FlemishBrabant Li
> mbourg Liege Namur WalloonBrabant Luxembourg SME publicfirm
note: industry35 omitted because of collinearity
Source | SS df MS Number of obs = 158
-------------+---------------------------------- F(45, 112) = 2.18
Model | 6.68837838 45 .148630631 Prob > F = 0.0005
Residual | 7.65031799 112 .068306411 R-squared = 0.4665
-------------+---------------------------------- Adj R-squared = 0.2521
Total | 14.3386964 157 .091329276 Root MSE = .26135
--------------------------------------------------------------------------------------
pctchangecarbonint~y | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------------------+----------------------------------------------------------------
firmsize | -.0578393 .0305485 -1.89 0.061 -.1183671 .0026886
profitability | 1.054629 .8352151 1.26 0.209 -.6002422 2.709501
leverage | .2631442 .1145173 2.30 0.023 .0362428 .4900456
age | -.0020298 .0010979 -1.85 0.067 -.0042052 .0001456
capitalintensity | .0978578 .1462035 0.67 0.505 -.1918257 .3875413
CAPEX | .4342122 .4407499 0.99 0.327 -.4390773 1.307502
KZindex | -.0204969 .0200661 -1.02 0.309 -.0602553 .0192616
elektricitygenerator | -.0338193 .2002871 -0.17 0.866 -.4306626 .3630239
Carbonleakage | .0798621 .0586428 1.36 0.176 -.036331 .1960553
industry10 | -.1052642 .1502458 -0.70 0.485 -.402957 .1924285
industry11 | -.4417411 .2124938 -2.08 0.040 -.8627704 -.0207118
industry13 | -.265114 .1980213 -1.34 0.183 -.6574678 .1272399
industry16 | -.4443672 .2416625 -1.84 0.069 -.9231906 .0344561
industry17 | -.1755922 .1746796 -1.01 0.317 -.5216974 .170513
industry19 | -.1436712 .186829 -0.77 0.444 -.513849 .2265066
industry20 | -.2736145 .1463656 -1.87 0.064 -.5636192 .0163902
industry21 | -.5619068 .2067879 -2.72 0.008 -.9716306 -.152183
industry22 | -.1453423 .1910228 -0.76 0.448 -.5238295 .233145
industry23 | -.3403978 .1449636 -2.35 0.021 -.6276246 -.0531711
industry24 | -.307153 .1580007 -1.94 0.054 -.6202112 .0059052
industry25 | -.3261937 .2527531 -1.29 0.200 -.8269916 .1746042
industry28 | -.2751739 .3498472 -0.79 0.433 -.9683514 .4180036
industry29 | -.397056 .2400379 -1.65 0.101 -.8726604 .0785485
industry30 | -.6808144 .3052009 -2.23 0.028 -1.285531 -.076098
industry35 | 0 (omitted)
industry42 | -.4649284 .1828208 -2.54 0.012 -.8271643 -.1026925
industry46 | -.3612973 .1671831 -2.16 0.033 -.6925492 -.0300455
industry47 | -.0231291 .3071088 -0.08 0.940 -.6316258 .5853676
industry49 | -.8118249 .2569667 -3.16 0.002 -1.320972 -.3026783
industry52 | -.3575171 .203461 -1.76 0.082 -.7606489 .0456147
industry63 | .412111 .3246491 1.27 0.207 -.2311395 1.055362
industry70 | -.3135855 .3382127 -0.93 0.356 -.9837105 .3565396
industry72 | -.7072208 .3071434 -2.30 0.023 -1.315786 -.0986554
industry81 | -.7269726 .3101123 -2.34 0.021 -1.34142 -.1125248
WestFlanders | .0873493 .1001563 0.87 0.385 -.1110974 .2857961
Hainaut | .1618636 .0947142 1.71 0.090 -.0258004 .3495275
Antwerp | .1570501 .0803885 1.95 0.053 -.0022294 .3163296
Brussels | .3587464 .1021681 3.51 0.001 .1563133 .5611794
FlemishBrabant | .0328263 .1434051 0.23 0.819 -.2513124 .316965
Limbourg | .0628833 .0888941 0.71 0.481 -.113249 .2390156
Liege | .1947639 .1069939 1.82 0.071 -.0172307 .4067585
Namur | .1066969 .3108182 0.34 0.732 -.5091495 .7225434
WalloonBrabant | .3392022 .1466473 2.31 0.023 .0486394 .6297649
Luxembourg | .1204264 .1682968 0.72 0.476 -.2130322 .453885
SME | -.2160014 .0870929 -2.48 0.015 -.3885649 -.0434379
publicfirm | .1041803 .1565156 0.67 0.507 -.2059354 .4142959
_cons | 1.076628 .6076935 1.77 0.079 -.1274386 2.280695
--------------------------------------------------------------------------------------
.
end of do-file
. estat imtest, white
White's test for Ho: homoskedasticity
against Ha: unrestricted heteroskedasticity
chi2(157) = 158.00
Prob > chi2 = 0.4626
Cameron & Trivedi's decomposition of IM-test
---------------------------------------------------
Source | chi2 df p
---------------------+-----------------------------
Heteroskedasticity | 158.00 157 0.4626
Skewness | 28.06 45 0.9775
Kurtosis | 1.27 1 0.2592
---------------------+-----------------------------
Total | 187.33 203 0.7779
---------------------------------------------------
. estat hettest
Breusch-Pagan / Cook-Weisberg test for heteroskedasticity
Ho: Constant variance
Variables: fitted values of pctchangecarbonintensity
chi2(1) = 10.03
Prob > chi2 = 0.0015Timea De Wispelaere
0 Response to heteroscedasticity test white vs BP
Post a Comment