Re: st: uncertain explanatory variable

[David Jacobs <jacobs.184@sociology.osu.edu>]

Check out the stata command eivreg or in English "errors in variables 
regression"  It lets you assign reliabilities to one or more explanatory 
variables and see what happens to the other coefficients in a model.

Dave Jacobs

At 12:14 PM 10/15/2003 +0100, you wrote:

> Dear All,
>
> I have a very general question, which I am hoping there is a standard 
> answer to.
>
> I am hoping to fit regression models (with both binary and continuous 
> outcomes), but for some of the individuals there is uncertainty (which is 
> quantified) in their explanatory variable, for example individual i may be 
> x1 wp. 0.5 x2 wp 0.25 and x3 wp 0.25. I wonder if there is a standard way 
> to deal with this, my initial thought was to incorportate the 
> probabilities with some kind of weighting strategy (maybe some adaptation 
> of fweights or pweights in stata), but I am not sure if this is a common 
> problem an/or solution.
>
> Any help or suggested references would be greatly appreciated.
>
> Kindest regards and many thanks
> Emma
>
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/