label define not showing first value for regression

[Marcel Meyer]

Hello,
First, I hope everyone had/has a happy easter this year !
I have a problem with my regression.
I first "grouped" some variables as following for example:

. generate neuro=0 if Neurocitism < mean_neurocitism
. replace neuro=1 if Neurocitism >= mean_neurocitism
. label define neuro 0 "low_neuro" 1 "high_neuro"
. label val neuro neuro

. generate extravert = 0 if Extraversion < mean_extraversion
. replace extravert = 1 if Extraversion >= mean_extraversion
. label define extravert 0 "low_extr" 1 "high_extr"
. label val extravert extravert

After that I tried to do a regression with interaction effects with:

. regress SWB i.neuro##i.extravert

The problem is it doesn't show the effects of low_neuro or low_extra

Source | SS df MS Number of obs = 21000
-------------+------------------------------ F( 3, 20996) = 11.82
Model | 99.005519 3 33.0018397 Prob > F = 0.0000
Residual | 58638.5647 20996 2.79284458 R-squared = 0.0017
-------------+------------------------------ Adj R-squared = 0.0015
Total | 58737.5702 20999 2.79716035 Root MSE = 1.6712

---------------------------------------------------------------------------------------
SWB | Coef. Std. Err. t P>|t| [95% Conf. Interval]
----------------------+----------------------------------------------------------------
neuro |
high_neuro | -.1988148 .0679501 -2.93 0.003 -.3320022 -.0656275
|
extravert |
high_extr | .1692158 .0672993 2.51 0.012 .037304 .3011277
|
neuro#extravert |
high_neuro#high_extr | .0140056 .0815663 0.17 0.864 -.1458707 .1738819
|
_cons | 6.928639 .0515492 134.41 0.000 6.827599 7.02968

The problem happens to every variable I use, it always excludes the first value, so it probably has to something with . label define or . label val but
I can't find out what it is.

I hope someone can (Probably, because I hope it will be a minor problem again) help me with that.
Thanks a lot in advance and kind regards
Marcel

[Marcos Almeida]

Hello Marcel,

Welcome to the Stata Forum / Statalist

Please read the FAQ, particularly on how to shate command and output. Thanks.

With regards to your question, I believe there is nothing wrong with the output.

It just presents the coefficient of the level 1 for each categorical variable, compared to the reference level (=0).

This is something to be expected overall.

[Carlo Lazzaro]

Marcel:
you you re-type -regress- as follows:

regress SWB i.neuro##i.extravert, basel

you may appreciate what's going on.

PS: Crossed in the cyberspace with Marcos' helpful reply.

[Marcel Meyer]

Thank you very much, Marcos and Carlo !
I tried to improve the my topic by putting the relevant information in "[ Code][/ Code]", hope it is now conform to the rules !
I think I understand what you both mean.
My problem is that I would like to also show the influence of the level 0 on the dependent variable,
meaning I'd also like to know the coefficient of "low_neuro" for example. Maybe I explained it wrongly.
Again, thank you a lot for the fast and understandable help, you're really improving my day !

Kind regards,
Marcel

[Carlo Lazzaro]

Marcel.
as per dummy-trap (please, see: https://en.wikipedia.org/wiki/Dummy_...le_(statistics) you simply cannot get both coefficients for level 0 and 1.
You can have only one of them at time, by playing around with the reference category in -i.<whatyoulike>.

[Marcos Almeida]

My problem is that I would like to also show the influence of the level 0 on the dependent variable,

You can use - margins - and - marginsplot - command for that.

[Marcel Meyer]

I am really sorry I didn't answer !!!
Thank you both very much for your help, it was really useful.
Kind regards,
Marcel