I forwarded this message to Patrick Royston, who
is not a member of Statalist. Below is his reply:
Nick
[email protected]
Patrick Royston
Suggestion for dealing with this problem:
1. Load the offending dataset into memory. I assume you have a variable
called, say, _id which defines the observation number. That is, the data are
organised as imputation 1 (obs 1,...,N), imputation 2 (obs 1,...,N), ... The
variable _id takes the values 1,...,N successively, for each of the m
imputations. If you don't have such a variable, please create one using the
egen command with the seq() fcn.
2. Type
. char _dta[mi_id] _i
3. Save the modified data.
Now micombine should be happy.
richard rogers wrote:
Greetings. I am trying to use micombine, part of the multiple imputation
features mentioned recently in 4-3. When I run the program, I get the
following error message: "no row-identifier variable found -- data may have
incorrect format".
Context: I am using the Federal Reserve Survey of Consumer Finances. The
imputed values are already calculated, and I have a variable that tracks the
imputation (number 1 through 5). The variable is a numeric format, but
given the error about formatting, I converted it to integer and string
formats, but still no luck. I am in Stata/SE 8.2. I have until this point
been using the Stata macro from the survey codebook, but micombine gives me
more functionality.
I am assuming that there is some simple trick here that I'm missing, or else
I completely don't understand this at all. I looked at the ado file and
found where the error message was being generated, but I'm not a real
programmer and much of what I saw was beyond my skill set.
*
* 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/
