Hello,
I'm posting this message because I have some questions about my data.
I have about 3,200 data spread over 29 different countries. I want to show the impact of culture on my dependent variable (I also integrate interactions).
I use the vce(robust) option to have standard errors.
When I run my regressions for the full sample, everything looks good (low VIF, prob>F = 0.0000, non-significant omitted variables test, etc.). The problem is that my data linked to the USA represents 40% of the sample, so for robustness analysis, I remove this data and run my regressions again.
This time I no longer have the prob>F (I have "."). The same applies if I remove the data for Canada (which only represents 8% of the database).
Here is the command :
As you can see, I include indicator variables for the year and the company's size. Are my results compromised because of this? Knowing that if I remove the vce(robust) option I see my prob>F statistic (which has gone from 0.0000 in the full sample to 0.0600 in the reduced sample).
I appreciate any help you can provide.
ps: I'll show you the two regression results with the full sample and the reduced sample
Full sample:
Reduced sample:
I'm posting this message because I have some questions about my data.
I have about 3,200 data spread over 29 different countries. I want to show the impact of culture on my dependent variable (I also integrate interactions).
I use the vce(robust) option to have standard errors.
When I run my regressions for the full sample, everything looks good (low VIF, prob>F = 0.0000, non-significant omitted variables test, etc.). The problem is that my data linked to the USA represents 40% of the sample, so for robustness analysis, I remove this data and run my regressions again.
This time I no longer have the prob>F (I have "."). The same applies if I remove the data for Canada (which only represents 8% of the database).
Here is the command :
Code:
regress car55_w DGEN centPERF c.centPERF#i.DGEN c.QUOTA#i.DGEN c.centSIZE#i.DGEN i.DGEN#c.centBWOMEN AGE centSIZE INDEP centBWOMEN DUALITY DEBTRATIO MARKETCAP_w BM_w ROA_w LNCommonSharesOutstanding_w UNEMPL POPGROWTH CORRUP INTEGRATION GDP LEGAL QUOTAS PHD CA MBA MA LAW EXP BUSY ln_AGEXE i.Year i.FESIZE2, vce(robust)
I appreciate any help you can provide.
ps: I'll show you the two regression results with the full sample and the reduced sample
Full sample:
Code:
Linear regression Number of obs = 3,196
F(55, 3140) = 95.08
Prob > F = 0.0000
R-squared = 0.0295
Root MSE = .06839
---------------------------------------------------------------------------------------------
| Robust
car55_w | Coefficient std. err. t P>|t| [95% conf. interval]
----------------------------+----------------------------------------------------------------
DGEN | -.0007034 .0029022 -0.24 0.809 -.0063939 .0049871
centFUTUR | -.0007972 .0071607 -0.11 0.911 -.0148373 .013243
|
DGEN#c.centFUTUR |
1 | .0448187 .0154657 2.90 0.004 .0144948 .0751427
|
DGEN#c.QUOTAS |
1 | .0288507 .0111366 2.59 0.010 .007015 .0506863
|
DGEN#c.centSIZE |
1 | -.0019211 .0009756 -1.97 0.049 -.0038341 -8.17e-06
|
DGEN#c.centBWOMEN |
1 | .0454247 .0290884 1.56 0.118 -.0116096 .102459
|
AGE | -.0013216 .0013604 -0.97 0.331 -.0039891 .0013458
centSIZE | -.0003165 .0005469 -0.58 0.563 -.0013889 .000756
INDEP | -.0057062 .0075767 -0.75 0.451 -.020562 .0091495
centBWOMEN | -.0137267 .0150495 -0.91 0.362 -.0432346 .0157812
DUALITY | .0039083 .0027764 1.41 0.159 -.0015354 .0093521
DEBTRATIO | .0044042 .00883 0.50 0.618 -.0129089 .0217173
MARKETCAP_w | -.0012352 .0015185 -0.81 0.416 -.0042125 .0017422
BM_w | -.002517 .0028371 -0.89 0.375 -.0080797 .0030458
ROA_w | -.0345445 .0175011 -1.97 0.048 -.0688593 -.0002297
LNCommonSharesOutstanding_w | .0009978 .0013187 0.76 0.449 -.0015877 .0035834
UNEMPL | .0003852 .0003876 0.99 0.320 -.0003746 .0011451
POPGROWTH | .0055129 .0037465 1.47 0.141 -.0018329 .0128587
CORRUP | .0034896 .0028474 1.23 0.220 -.0020933 .0090725
INTEGRATION | .0047318 .0029201 1.62 0.105 -.0009937 .0104573
GDP | .0030491 .0016574 1.84 0.066 -.0002005 .0062987
LEGAL | -.0038553 .0053843 -0.72 0.474 -.0144124 .0067017
QUOTAS | .0004738 .0063233 0.07 0.940 -.0119244 .0128719
PHD | .0013761 .0040603 0.34 0.735 -.006585 .0093372
CA | -.0194509 .0130981 -1.49 0.138 -.0451327 .0062309
MBA | -.0008793 .0029929 -0.29 0.769 -.0067476 .004989
MA | -.002767 .0036924 -0.75 0.454 -.0100067 .0044728
LAW | -.0046762 .005536 -0.84 0.398 -.0155307 .0061784
EXP | -.0072091 .0033822 -2.13 0.033 -.0138405 -.0005776
BUSY | .0034211 .0033058 1.03 0.301 -.0030606 .0099028
ln_AGEXE | .0131108 .0092635 1.42 0.157 -.0050524 .031274
|
Year |
2003 | .0159833 .0194859 0.82 0.412 -.022223 .0541897
2004 | -.0161759 .015624 -1.04 0.301 -.0468103 .0144585
2005 | -.0009112 .0162589 -0.06 0.955 -.0327903 .030968
2006 | -.0013393 .0141565 -0.09 0.925 -.0290963 .0264177
2007 | -.0026169 .014224 -0.18 0.854 -.0305062 .0252723
2008 | .0010135 .0149915 0.07 0.946 -.0283807 .0304078
2009 | -.0028215 .0144292 -0.20 0.845 -.0311131 .0254702
2010 | -.0093062 .0131572 -0.71 0.479 -.0351038 .0164913
2011 | -.0169854 .0133098 -1.28 0.202 -.0430821 .0091114
2012 | -.0148416 .0134009 -1.11 0.268 -.0411169 .0114338
2013 | -.0110633 .0128892 -0.86 0.391 -.0363354 .0142088
2014 | -.0096998 .0130592 -0.74 0.458 -.0353051 .0159056
2015 | -.0053336 .0127978 -0.42 0.677 -.0304265 .0197592
2016 | -.0089414 .0127195 -0.70 0.482 -.0338807 .015998
2017 | -.0074988 .0127983 -0.59 0.558 -.0325926 .017595
2018 | -.0194639 .0130291 -1.49 0.135 -.0450104 .0060826
2019 | -.0054806 .0133076 -0.41 0.680 -.0315729 .0206118
|
FESIZE2 |
2 | .2563835 .0225849 11.35 0.000 .2121009 .300666
3 | .2348571 .0101594 23.12 0.000 .2149374 .2547767
4 | .2337393 .0106437 21.96 0.000 .2128699 .2546086
5 | .2408719 .0118452 20.33 0.000 .2176468 .2640971
6 | .2411293 .0137035 17.60 0.000 .2142606 .2679979
7 | .2416817 .0169828 14.23 0.000 .2083832 .2749801
8 | .2505442 .020197 12.41 0.000 .2109435 .290145
|
_cons | -.3608112 .0730594 -4.94 0.000 -.5040603 -.2175622
Reduced sample:
Code:
Linear regression Number of obs = 1,593
F(53, 1537) = .
Prob > F = .
R-squared = 0.0443
Root MSE = .07167
---------------------------------------------------------------------------------------------
| Robust
car55_w | Coefficient std. err. t P>|t| [95% conf. interval]
----------------------------+----------------------------------------------------------------
DGEN | .0067783 .005162 1.31 0.189 -.003347 .0169035
centFUTUR | .0000168 .0072818 0.00 0.998 -.0142664 .0143001
|
DGEN#c.centFUTUR |
1 | .0340473 .0167192 2.04 0.042 .0012524 .0668423
|
DGEN#c.QUOTAS |
1 | .0226001 .012364 1.83 0.068 -.0016519 .0468522
|
DGEN#c.centSIZE |
1 | -.0016121 .0012494 -1.29 0.197 -.0040627 .0008386
|
DGEN#c.centBWOMEN |
1 | .0314506 .0435058 0.72 0.470 -.0538865 .1167876
|
AGE | -.0040321 .0019804 -2.04 0.042 -.0079167 -.0001475
centSIZE | -.0005461 .000653 -0.84 0.403 -.0018269 .0007347
INDEP | -.0074823 .0095797 -0.78 0.435 -.0262729 .0113084
centBWOMEN | -.0040372 .0201498 -0.20 0.841 -.0435611 .0354868
DUALITY | .0043646 .0044447 0.98 0.326 -.0043537 .013083
DEBTRATIO | .0109233 .0132116 0.83 0.408 -.0149913 .0368379
MARKETCAP_w | -.0027462 .0020072 -1.37 0.171 -.0066832 .0011909
BM_w | -.0017192 .0033959 -0.51 0.613 -.0083803 .0049419
ROA_w | .0027042 .0275708 0.10 0.922 -.0513762 .0567845
LNCommonSharesOutstanding_w | -.0006544 .0016692 -0.39 0.695 -.0039285 .0026197
UNEMPL | .0003371 .0004319 0.78 0.435 -.0005101 .0011842
POPGROWTH | .0042424 .0039422 1.08 0.282 -.0034904 .0119751
CORRUP | .0021214 .0033061 0.64 0.521 -.0043636 .0086065
INTEGRATION | .0039104 .0032206 1.21 0.225 -.0024068 .0102276
GDP | .0028411 .0027489 1.03 0.302 -.002551 .0082332
LEGAL | -.0011298 .0054773 -0.21 0.837 -.0118735 .0096138
QUOTAS | .0001542 .0070254 0.02 0.982 -.0136263 .0139347
PHD | .0058123 .0058507 0.99 0.321 -.005664 .0172885
CA | -.0191117 .013741 -1.39 0.164 -.0460648 .0078414
MBA | .001409 .0046287 0.30 0.761 -.0076701 .0104881
MA | -.0005586 .0050308 -0.11 0.912 -.0104265 .0093093
LAW | -.0122426 .0097059 -1.26 0.207 -.0312809 .0067957
EXP | -.0126975 .0050337 -2.52 0.012 -.022571 -.0028239
BUSY | .0022522 .0049623 0.45 0.650 -.0074813 .0119857
ln_AGEXE | .0211569 .0138933 1.52 0.128 -.0060949 .0484087
|
Year |
2003 | -.0145246 .0292096 -0.50 0.619 -.0718194 .0427702
2004 | -.0338672 .01575 -2.15 0.032 -.0647608 -.0029735
2005 | -.0452805 .013995 -3.24 0.001 -.0727319 -.0178291
2006 | -.0304899 .0143438 -2.13 0.034 -.0586253 -.0023545
2007 | -.0346337 .0174599 -1.98 0.047 -.0688814 -.000386
2008 | -.0234888 .0170425 -1.38 0.168 -.0569179 .0099403
2009 | -.023938 .0155296 -1.54 0.123 -.0543994 .0065235
2010 | -.0324411 .0136871 -2.37 0.018 -.0592884 -.0055938
2011 | -.052807 .0138124 -3.82 0.000 -.0799002 -.0257139
2012 | -.0457398 .0141955 -3.22 0.001 -.0735844 -.0178953
2013 | -.0391303 .0124888 -3.13 0.002 -.0636271 -.0146335
2014 | -.0369942 .0129445 -2.86 0.004 -.062385 -.0116034
2015 | -.0387998 .0126482 -3.07 0.002 -.0636095 -.0139902
2016 | -.0357512 .0131008 -2.73 0.006 -.0614484 -.0100539
2017 | -.0369101 .0123544 -2.99 0.003 -.0611434 -.0126769
2018 | -.0532437 .0130374 -4.08 0.000 -.0788167 -.0276708
2019 | -.029239 .013521 -2.16 0.031 -.0557605 -.0027175
|
FESIZE2 |
2 | .2359234 .0778803 3.03 0.002 .0831606 .3886863
3 | .242945 .0160831 15.11 0.000 .2113978 .2744921
4 | .2434875 .0153902 15.82 0.000 .2132996 .2736754
5 | .2552508 .0169289 15.08 0.000 .2220447 .2884569
6 | .2562097 .019285 13.29 0.000 .218382 .2940374
7 | .2652543 .0216385 12.26 0.000 .2228102 .3076983
8 | .2817825 .0260609 10.81 0.000 .2306638 .3329012
|
_cons | -.2921799 .110172 -2.65 0.008 -.5082832 -.0760767
Comment