.

[Elene Timmermans]

.

[Jared Greathouse]

None, no difference.

[Andrew Musau]

No difference, except your commands are not equivalent:

reg hirinv $X_CC_nref $X_VC_nref, cluster (id_resp)

regress hirinv $X_CC_nref, cluster (id_resp)

ADDED IN EDIT: OP, please do not mangle your posts. The reply in #2 and above refer to the now deleted post in #1:

Difference between reg and regress

Today, 08:06
Hello everyone


I am confused about using 'reg' or 'regress' to run a linear regression. I am wondering what the differences is between the two in order to choose the right one for my analysis because I get slightly different result when running them.

If necessary, you can find the output of both commands below.


reg hirinv $X_CC_nref $X_VC_nref, cluster (id_resp)

Linear regression Number of obs = 870
F(17, 173) = 9.30
Prob > F = 0.0000
R-squared = 0.1483
Root MSE = 2.2636

(Std. err. adjusted for 174 clusters in id_resp)
----------------------------------------------------------------------------------------
| Robust
hirinv | Coefficient std. err. t P>|t| [95% conf. interval]
-----------------------+----------------------------------------------------------------
X_CC_Vrouw | -.0374627 .1012421 -0.37 0.712 -.2372914 .162366
X_CC_38±1jaar | -.080413 .1375089 -0.58 0.559 -.351824 .1909981
X_CC_48±1jaar | -.3967042 .154455 -2.57 0.011 -.701563 -.0918454
X_CC_1_kind | -.2386436 .1278716 -1.87 0.064 -.4910329 .0137457
X_CC_2_kinderen | -.2823243 .1627294 -1.73 0.085 -.6035149 .0388663
X_CC_tussen_20en40km | -.2397479 .137268 -1.75 0.082 -.5106836 .0311878
X_CC_meer_40km | -.9571473 .1714826 -5.58 0.000 -1.295615 -.6186798
X_CC_tussen_1en5j | .9600428 .1636607 5.87 0.000 .6370141 1.283072
X_CC_meer_5j | 1.433364 .1654897 8.66 0.000 1.106725 1.760003
X_CC_sportvereniging | .428354 .139563 3.07 0.002 .1528885 .7038196
X_CC_vrijwilliger | .3437873 .1389075 2.47 0.014 .0696157 .617959
X_CC_vastewerkuren | -.0506587 .180473 -0.28 0.779 -.4068711 .3055538
X_CC_flexibelewerkuren | .0366531 .1479573 0.25 0.805 -.2553808 .328687
X_CC_telewerk | -.400814 .1453688 -2.76 0.006 -.6877387 -.1138894
VC_hogeopleiding | -.3341092 .2773141 -1.20 0.230 -.8814638 .2132455
VC_hogesocialeinter | -.3457912 .298901 -1.16 0.249 -.9357533 .2441709
VC_gemfysiekeact | -.8568719 .3258443 -2.63 0.009 -1.500014 -.2137298
_cons | 7.307411 .2991675 24.43 0.000 6.716922 7.897899
----------------------------------------------------------------------------------------

. regress hirinv $X_CC_nref, cluster (id_resp)

Linear regression Number of obs = 870
F(14, 173) = 10.97
Prob > F = 0.0000
R-squared = 0.1114
Root MSE = 2.3081

(Std. err. adjusted for 174 clusters in id_resp)
----------------------------------------------------------------------------------------
| Robust
hirinv | Coefficient std. err. t P>|t| [95% conf. interval]
-----------------------+----------------------------------------------------------------
X_CC_Vrouw | -.0443988 .1001413 -0.44 0.658 -.2420549 .1532573
X_CC_38±1jaar | -.0680373 .1398626 -0.49 0.627 -.3440942 .2080195
X_CC_48±1jaar | -.3904702 .1546352 -2.53 0.012 -.6956847 -.0852556
X_CC_1_kind | -.2435841 .1264929 -1.93 0.056 -.4932521 .0060839
X_CC_2_kinderen | -.2925112 .1627436 -1.80 0.074 -.6137299 .0287075
X_CC_tussen_20en40km | -.232772 .1389181 -1.68 0.096 -.5069646 .0414205
X_CC_meer_40km | -.9485287 .1699266 -5.58 0.000 -1.283925 -.6131325
X_CC_tussen_1en5j | .9366292 .1638382 5.72 0.000 .6132501 1.260008
X_CC_meer_5j | 1.41874 .1659251 8.55 0.000 1.091242 1.746238
X_CC_sportvereniging | .4241242 .1407762 3.01 0.003 .1462642 .7019841
X_CC_vrijwilliger | .3534466 .1430508 2.47 0.014 .0710971 .6357961
X_CC_vastewerkuren | -.0624612 .1811294 -0.34 0.731 -.4199693 .2950469
X_CC_flexibelewerkuren | .0589444 .1470345 0.40 0.689 -.231268 .3491568
X_CC_telewerk | -.4150587 .146764 -2.83 0.005 -.7047373 -.12538
_cons | 6.690371 .2369725 28.23 0.000 6.222641 7.1581
----------------------------------------------------------------------------------------


Thank you in advance!

[Jared Greathouse]

I forgot to add that the code wasn't the same, but yes, that's the real reason, the code was different. For sake of reproducibility, which I won't delete, consider

sysuse auto, clear

reg price weight
regress price weight