Hello,
I am using 4 binomial models. Logit, xtlogit re, xtlogit pa, and xtlogit fe. After logit I use “estat classification” normally. After RE, FE, and PA I am trying use “estat classification” but show this error: “invalid subcommand classification”.
Can someone help me about how can I make classified after XT models?
Thanks
Erasmo

Code:
. logit DUMMYRET LNVM, vce (cluster id) nolog

Logistic regression                               Number of obs   =      21236
                                                  Wald chi2(1)    =      70.28
                                                  Prob > chi2     =     0.0000
Log pseudolikelihood = -14632.346                 Pseudo R2       =     0.0032

                                   (Std. Err. adjusted for 619 clusters in id)
------------------------------------------------------------------------------
             |               Robust
    DUMMYRET |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        LNVM |   .0578859   .0069048     8.38   0.000     .0443527    .0714191
       _cons |  -.6494957   .0960829    -6.76   0.000    -.8378146   -.4611768
------------------------------------------------------------------------------

. estat classification

Logistic model for DUMMYRET

              -------- True --------
Classified |         D            ~D  |      Total
-----------+--------------------------+-----------
     +     |      9488          7898  |      17386
     -     |      1779          2071  |       3850
-----------+--------------------------+-----------
   Total   |     11267          9969  |      21236

Classified + if predicted Pr(D) >= .5
True D defined as DUMMYRET != 0
--------------------------------------------------
Sensitivity                     Pr( +| D)   84.21%
Specificity                     Pr( -|~D)   20.77%
Positive predictive value       Pr( D| +)   54.57%
Negative predictive value       Pr(~D| -)   53.79%
--------------------------------------------------
False + rate for true ~D        Pr( +|~D)   79.23%
False - rate for true D         Pr( -| D)   15.79%
False + rate for classified +   Pr(~D| +)   45.43%
False - rate for classified -   Pr( D| -)   46.21%
--------------------------------------------------
Correctly classified                        54.43%
--------------------------------------------------

. xtlogit DUMMYRET LNVM, re vce (cluster id) nolog

Calculating robust standard errors:

Random-effects logistic regression              Number of obs      =     21236
Group variable: id                              Number of groups   =       619

Random effects u_i ~ Gaussian                   Obs per group: min =         1
                                                               avg =      34.3
                                                               max =        95

Integration method: mvaghermite                 Integration points =        12

                                                Wald chi2(1)       =     21.88
Log pseudolikelihood  = -14607.845              Prob > chi2        =    0.0000

                                   (Std. Err. adjusted for 619 clusters in id)
------------------------------------------------------------------------------
             |               Robust
    DUMMYRET |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        LNVM |   .0377137   .0077938     4.84   0.000     .0224381    .0529892
       _cons |  -.3800981   .1068064    -3.56   0.000    -.5894348   -.1707615
-------------+----------------------------------------------------------------
    /lnsig2u |  -2.946617   .2009394                     -3.340451   -2.552783
-------------+----------------------------------------------------------------
     sigma_u |   .2291661   .0230242                      .1882047    .2790425
         rho |   .0157125   .0031076                       .010652    .0231208
------------------------------------------------------------------------------

. estat classification
invalid subcommand classification
r(321);