Re: st: Interpretation of nldecompose regression output

[Francisco Rowe <frowe@ucn.cl>]

Hi Ilaria,

Please refer to:
Powers, D, Yoshioka, H and Yun, M forthcoming, 'mvdcmp: Multivariate decomposition for nonlinear response models', The Stata Journal, pp. 1-20.

The authors provide, to me, a fairly clear explanations of decomposition outcomes.

Cheers,

Francisco.

On 13/10/2011, at 3:15 AM, Ilaria Maselli wrote:

> Hello,
> I am using nldecompose to apply a Blinder-Oaxaca decomposition for ologit.
> 
> Unfortunately I do not understand how to interpret the following
> regression output.
> Any help is highly appreciated.  Many thanks in advance,
> Ilaria
> 
> nldecompose, by(routine) : ologit ef5 isced q2c  exper_sq if countcod==4
> [pweight= w4]
> 
>                                                   Number of obs (A) =
> 87
>                                                   Number of obs (B) =
> 605
> 
> --------------------------------------------------------------------------
> ----
>      Results |      Coef.  Percentage
> --------------+-----------------------------------------------------------
> ----
> Omega = 1    |
>         Char |  -.4017469    24.7564%
>         Coef |  -1.221053    75.2436%
> --------------+-----------------------------------------------------------
> ----
> Omega = 0    |
>         Char |   .4488521  -27.65912%
>         Coef |  -2.071652   127.6591%
> --------------+-----------------------------------------------------------
> ----
>          Raw |    -1.6228        100%
> --------------------------------------------------------------------------
> ----
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/