Re: st: the impute command

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

At 10:56 AM 5/19/2004 -0400, Robert Bozick wrote:
> Hello -
>
> I have a few questions concerning the impute command.
>
> I am trying to impute values to a dichotomous variable coded 1 if 
> individuals were employed and 0 if they were not.
>
> First, does the impute command treat all the predictor variables in the 
> same way that a regression command does?  More specifically, if I am using 
> a set of dummy variables, do I omit one category?  I ask this because I 
> was doing some test runs and I included ALL variables in the impute 
> command and it did not give me the message "variable was dropped due to 
> collinearity."
It doesn't matter.  It will basically drop a variable for you, even though 
it does not say so.


> Second, is there a way to specify the impute command to round imputed 
> values either to 1 or 0 depending on the prediction (e.g. predicted values 
> 0 - .49 become 0 and values .50-1 become 1)  ?
You could just follow the impute command with a recode command.  Keep in 
mind that the imputed values could be less than 0 or greater than 1.  So, 
something like

Recode y (min/.50 = 0)(.50/max = 1)


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc

*
*   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/