Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: imputing data values using hotdeck
From
Richard Williams <[email protected]>
To
[email protected], "[email protected]" <[email protected]>
Subject
Re: st: imputing data values using hotdeck
Date
Thu, 09 Feb 2012 17:03:51 -0500
At 12:59 PM 2/9/2012, Krohmer, Anne (MDH) wrote:
Hi, I am working with STATA version 10 and I have some problems
using the hotdeck imputation command. I am using survey data and
several respondents (~100) were not asked a question that they
actually should have been asked so I need to impute these values.
However, this variable also has about 10,000 other missing values.
How do I impute the variables just for these 100 cases and not
impute the other missing cases? I have a variable that flags which
cases these are. Any thoughts? Our resident Stata expert is out of
the office today. Thanks!
Assuming we are talking about the same command (hotdeck is user
written, right?), hotdeck has an -if- option for selecting cases. So,
I assume you could use that to exclude the 10,000 other cases.
As a sidelight, I just noticed that Stata 12 distinguishes between
"soft" missing and "hard" missing, where hard missing is .a, .b, etc.
The mi command does not impute values for cases with hard missing.
That sounds like it would be very helpful for cases like this, or for
when you want to treat "not applicable" differently from "No Answer."
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
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/