RE: st: imputing continuous values when respondents select categories, e.g., income category

[Richard Williams <Richard.A.Williams.5@ND.edu>]

At 12:27 AM 4/25/2009, Roy Wada wrote:
> Bootstrap will correct the standard error.

Bootstrapping being yet another weak area of mine, I suppose that 
could be true.  But, does that mean that if you bootstrapped, single 
imputation of missing data would be ok?  Does it mean that regardless 
of whether you have the real value of income or the estimated value, 
either way the standard errors will be computed correctly?

> It should not matter how E(y|x,z) is calcualted, as long as the
> conditions (i.e. exclusions, non-weak instruments) are met.

It is similar to the problems you have with 2SLS.  If x1, x2 and x3 
are used to estimate income, then at least one of those variables has 
to be excluded from models that include estimated income as an 
independent variable.

Interesting discussion.  Maybe using intreg in this way is a good 
idea, but I'd feel more comfortable if there were citations justifying it.


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW:    http://www.nd.edu/~rwilliam

*
*   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/