re: st: Using Post-Estimation Results for further Calculations

[Christopher Baum <kit.baum@bc.edu>]

<>
I wonder, whether I can use results of post-estimation calculations 
without copy and paste for further things. Especially, when I run a 
normal regression with the reg command and check for multicollinearity, 
I use the vif command. However, is it possible to store the result of 
vif, so that I can use it, when I need to?

After running a regression, do -estat vif- and -return list-:

. qui reg price mpg weight turn length

. estat vif

    Variable |       VIF       1/VIF  
-------------+----------------------
      weight |     10.86    0.092065
      length |     10.83    0.092344
        turn |      4.21    0.237598
         mpg |      2.96    0.338118
-------------+----------------------
    Mean VIF |      7.21

. ret li

scalars:
              r(vif_4) =  2.957543134689331
              r(vif_3) =  4.208797454833984
              r(vif_2) =  10.82910823822021
              r(vif_1) =  10.86192989349365

macros:
             r(name_4) : "mpg"
             r(name_3) : "turn"
             r(name_2) : "length"
             r(name_1) : "weight"

Everything (except the mean vif) is accessible, and you can calc that:

. scalar mean=0.25*(r(vif_1)+r(vif_2)+r(vif_3)+r(vif_4))

. di mean
7.2143447


Kit

Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html


*
*   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/