Quantcast
Channel: Statalist
Viewing all articles
Browse latest Browse all 73275

-mlogit- bug with labels?

$
0
0
Here's my dependent variable:
Code:
. tab catvar

Would you recommend                     |      Freq.     Percent        Cum.
----------------------------------------+-----------------------------------
          1 Yes I strongly recommend it |        574       65.08       65.08
                                      2 |        166       18.82       83.90
                                      3 |         44        4.99       88.89
4 Not sure whether to recommend it or n |         59        6.69       95.58
                                      5 |         18        2.04       97.62
                                      6 |          6        0.68       98.30
   7 No I strongly recommend against it |         15        1.70      100.00
----------------------------------------+-----------------------------------
                                  Total |        882      100.00


Here's the problem:
Code:
. mlogit catvar group

7_No_I_strongly_recommend_agains invalid name
r(198);

Here's the "solution":
Code:
. label drop catvarlabel
. tab catvar

Would you   |
  recommend |      Freq.     Percent        Cum.
------------+-----------------------------------
          1 |        574       65.08       65.08
          2 |        166       18.82       83.90
          3 |         44        4.99       88.89
          4 |         59        6.69       95.58
          5 |         18        2.04       97.62
          6 |          6        0.68       98.30
          7 |         15        1.70      100.00
------------+-----------------------------------
      Total |        882      100.00


. mlogit catvar group

Multinomial logistic regression                 Number of obs     =        882
                                                LR chi2(6)        =      12.79
                                                Prob > chi2       =     0.0465
Log likelihood = -970.02327                     Pseudo R2         =     0.0065

------------------------------------------------------------------------------
catvar |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
1            |  (base outcome)
-------------+----------------------------------------------------------------
2            |
       group |  -.1186057   .1763223    -0.67   0.501    -.4641912    .2269797
       _cons |  -1.179581   .1255066    -9.40   0.000     -1.42557    -.933593
-------------+----------------------------------------------------------------
3            |
       group |  -.8807414   .3342974    -2.63   0.008    -1.535952   -.2255306
       _cons |  -2.197224   .1924501   -11.42   0.000    -2.574419   -1.820029
-------------+----------------------------------------------------------------
4            |
       group |  -.6384811   .2819127    -2.26   0.024     -1.19102   -.0859425
       _cons |  -1.987504   .1753018   -11.34   0.000    -2.331089   -1.643919
-------------+----------------------------------------------------------------
5            |
       group |  -.5705909   .4906724    -1.16   0.245    -1.532291    .3911094
       _cons |  -3.200527   .3075919   -10.41   0.000    -3.803396   -2.597658
-------------+----------------------------------------------------------------
6            |
       group |  -.1186057   .8207679    -0.14   0.885    -1.727281     1.49007
       _cons |   -4.49981   .5805489    -7.75   0.000    -5.637665   -3.361955
-------------+----------------------------------------------------------------
7            |
       group |   .0149257   .5242617     0.03   0.977    -1.012608     1.04246
       _cons |  -3.652512   .3828327    -9.54   0.000     -4.40285   -2.902174
------------------------------------------------------------------------------
But that is not very satisfying. I have other variables in this dataset with the same label, but -mlogit- works fine. The labels are created by a do-file, not imported from another format. Why the strange error?








Viewing all articles
Browse latest Browse all 73275

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>