Regression issues

[Rodrigo Soares]

Dear all,

I'm sorry for my elementary question, but if I have a regression with some dummies, how do I know which is the reference category (because it seems senseless to set all the time, e.g., if female==0).

reg lwage educ exper expersq occupa union indus female black

Many thanks.

[Carlo Lazzaro]

Rodrigo:
in a previous today's post William Lisowski wisely pointed you out to the FAQ to increase the effectiveness of your postings. Please take a look. Thanks.
That said, if you set female=0, you're limiting the effect of that predictor to one of its level only.
If you want to choose the reference category of your categorical variable(s), please take a look at -fvvarlist- notation helpfile.
As you can see in the following toy-example, switching the reference category is easy and invoking the -allbaselevels- option gives you the exact perception of what Stata is doing behind the curtain:

. sysuse auto.dta
(1978 automobile data)

. regress price i.foreign, allbaselevels

      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(1, 72)        =      0.17
       Model |  1507382.66         1  1507382.66   Prob > F        =    0.6802
    Residual |   633558013        72  8799416.85   R-squared       =    0.0024
-------------+----------------------------------   Adj R-squared   =   -0.0115
       Total |   635065396        73  8699525.97   Root MSE        =    2966.4

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
     foreign |
   Domestic  |          0  (base)
    Foreign  |   312.2587   754.4488     0.41   0.680    -1191.708    1816.225
             |
       _cons |   6072.423    411.363    14.76   0.000     5252.386     6892.46
------------------------------------------------------------------------------

. regress price ib1.foreign, allbaselevels

      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(1, 72)        =      0.17
       Model |  1507382.66         1  1507382.66   Prob > F        =    0.6802
    Residual |   633558013        72  8799416.85   R-squared       =    0.0024
-------------+----------------------------------   Adj R-squared   =   -0.0115
       Total |   635065396        73  8699525.97   Root MSE        =    2966.4

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
     foreign |
   Domestic  |  -312.2587   754.4488    -0.41   0.680    -1816.225    1191.708
    Foreign  |          0  (base)
             |
       _cons |   6384.682   632.4346    10.10   0.000     5123.947    7645.417
------------------------------------------------------------------------------

.

In addition, getting yourself familiar with -fvvarlist- notation avoids creating interaction by hand, as in your code above:

exper expersq

Please note that when you ignore -fvvarlist- notation you cannot exploit the wonderful capabiities of -margins- and -marginsplot-, because Stata consider -exper- and -expersq- as two different variables.
The most convenient notation for interactions is:

c.exper##c.exper