# st: Predicted Probability after xtgee (urgent!!!)

 From ckang2@gmail.com To statalist Subject st: Predicted Probability after xtgee (urgent!!!) Date Thu, 19 Jul 2007 01:39:18 -0400

```Dear STATA listers

Could you anyone teach me how to get predicted probability after running xtgee?

With searching the archives, some reconmmend to use mfx.
But since I'm beginner of STATA, I could not figure out and interpret
the results, and it looks more like odd ratio, not predicted
probability.

For example, these are the results actually I got by running mfx after xtgee.

arginal effects after xtgee
y  = xb (predict)
=  .05777175
------------------------------------------------------------------------------
variable |      dy/dx    Std. Err.     z    P>|z|  [    95% C.I.   ]      X
---------+--------------------------------------------------------------------
x1        |    .002552      .00662    0.39   0.700  -.010432  .015536   .580848
x2        |   1.88e-09      .00000    1.79   0.073  -1.8e-10  3.9e-09    304447
x3        |  -.0156874      .00523   -3.00   0.003   -.02593 -.005445   .564654
x4        |  -.0145082      .00677   -2.14   0.032  -.027772 -.001244   .332159
x5        |  -.0227091      .00653   -3.48   0.001  -.035508  -.00991   .180099
x6        |   .5816271       .0339   17.16   0.000   .515188  .648067    .02012
x7        |   .0449566      .01105    4.07   0.000   .023306  .066608   .380295
x8        |   .1685375      .02724    6.19   0.000   .115158  .221917   .016841
x9        |   .0713164      .01873    3.81   0.000   .034603   .10803   .220498
x10      |  -.0024053       .0095   -0.25   0.800  -.021029  .016218   .550549
------------------------------------------------------------------------------
(*) dy/dx is for discrete change of dummy variable from 0 to 1

My quesitons here are three.

1) how is it possible that x2's dy/dx is over 1? I think dy.dx is odd
ratio. Am I right?
2) what does "X" in the last column of the first raw mean?
3) is there any other way (STATA command like "prvalue" or "estsimp")
to get predicted probability varying across the change of x (I.v.)?