Reference category in logistic regression??

[Cathrine Erlandsen]

I have a variable where 0 is coded for X, 1 for Y, 2 for Z... etc. (numeric variable). How do I then vary between reference category when running a logistic regression to get result for each category?

Whats the code?


Im sorry if the questions is a bit stupid but Im totally new to this.. and my googling have failed so far!

[Andrew Musau]

Choose one base category and then use the test command for comparisons of non base categories. Otherwise, use ibn.catvar to choose a base, e.g., ib0.catvar for 0, ib1.catvar for 1, etc. See

help fvvarlist

webuse lbw, clear
tab race, nolab
logit low age smoke ib1.race
test 2.race= 3.race
logit low age smoke ib2.race

Res.:

. tab race, nolab

       race |      Freq.     Percent        Cum.
------------+-----------------------------------
          1 |         96       50.79       50.79
          2 |         26       13.76       64.55
          3 |         67       35.45      100.00
------------+-----------------------------------
      Total |        189      100.00

.
. logit low age smoke ib1.race

Iteration 0:   log likelihood =   -117.336  
Iteration 1:   log likelihood = -109.57893  
Iteration 2:   log likelihood = -109.43115  
Iteration 3:   log likelihood =  -109.4311  
Iteration 4:   log likelihood =  -109.4311  

Logistic regression                             Number of obs     =        189
                                                LR chi2(4)        =      15.81
                                                Prob > chi2       =     0.0033
Log likelihood =  -109.4311                     Pseudo R2         =     0.0674

------------------------------------------------------------------------------
         low |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         age |  -.0348828   .0334024    -1.04   0.296    -.1003502    .0305847
       smoke |    1.10055   .3719453     2.96   0.003     .3715511     1.82955
             |
        race |
      black  |   1.011413   .4934234     2.05   0.040     .0443209    1.978505
      other  |    1.05673   .4059583     2.60   0.009     .2610665    1.852394
             |
       _cons |  -1.007554   .8616628    -1.17   0.242    -2.696382    .6812744
------------------------------------------------------------------------------

.
. test 2.race= 3.race

 ( 1)  [low]2.race - [low]3.race = 0

           chi2(  1) =    0.01
         Prob > chi2 =    0.9265

.
. logit low age smoke ib2.race

Iteration 0:   log likelihood =   -117.336  
Iteration 1:   log likelihood = -109.57893  
Iteration 2:   log likelihood = -109.43115  
Iteration 3:   log likelihood =  -109.4311  
Iteration 4:   log likelihood =  -109.4311  

Logistic regression                             Number of obs     =        189
                                                LR chi2(4)        =      15.81
                                                Prob > chi2       =     0.0033
Log likelihood =  -109.4311                     Pseudo R2         =     0.0674

------------------------------------------------------------------------------
         low |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         age |  -.0348828   .0334024    -1.04   0.296    -.1003502    .0305847
       smoke |    1.10055   .3719453     2.96   0.003     .3715511     1.82955
             |
        race |
      white  |  -1.011413   .4934234    -2.05   0.040    -1.978505   -.0443209
      other  |   .0453171   .4909151     0.09   0.926    -.9168588    1.007493
             |
       _cons |   .0038592   .8443829     0.00   0.996    -1.651101    1.658819
------------------------------------------------------------------------------

.

[Carlo Lazzaro]

Cathrine:
welcome to this forum.
This is not a stupid question at all.
The following toy-example can be helpful:

. sysuse auto.dta
. logit foreign i.rep78 if rep78>=3

Iteration 0:   log likelihood = -38.411464
Iteration 1:   log likelihood = -27.676628
Iteration 2:   log likelihood = -27.446054
Iteration 3:   log likelihood = -27.444671
Iteration 4:   log likelihood = -27.444671

Logistic regression                             Number of obs     =         59
                                                LR chi2(2)        =      21.93
                                                Prob > chi2       =     0.0000
Log likelihood = -27.444671                     Pseudo R2         =     0.2855

------------------------------------------------------------------------------
     foreign |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
       rep78 |
          4  |   2.197225   .7698004     2.85   0.004     .6884436    3.706006
          5  |   3.701302   .9906975     3.74   0.000     1.759571    5.643033
             |
       _cons |  -2.197225   .6085806    -3.61   0.000    -3.390021   -1.004428
------------------------------------------------------------------------------

.
predict fitted, xb

///To be compared to the fitted values calculated by hand///
/// -2.197225 if rep78==3///
///. di -2.197225 + 2.197225///
///0 if rep78==4///
///. di -2.197225 + 3.701302///
///1.504077 if rep78==5///

.

See -fvvarlist- notation for more details about creating categorical variables and interactions.

PS: crossed in the cyber-space with Andrew's helpdful reply,that focuses on a different topic.