Hi Mouna,
When you specify ords = "X5", you are saying that X5 is an ordinal variable
and so the imputation are constrained to only take values that are observed
in the data. An alternative is to use the "bounds" argument to constrain
the imputations to be within certain bounds. See Section 4.6.3 of the
Amelia manual here:
http://r.iq.harvard.edu/docs/amelia/amelia.pdf
Cheers,
Matt
~~~~~~~~~~~
Matthew Blackwell
Assistant Professor of Government
Harvard University
url:
http://www.mattblackwell.org
On Fri, Aug 29, 2014 at 9:10 AM, mouna kessentini <kessentini_mouna(a)yahoo.fr
wrote:
Dear professors,
I am a Phd student at the University of Paris VIII and I am working on
financial issue in a panel data. I've come to use the multiple imputation
for my missing data and I've been very interested in your method.
Since I was working with R, I installed your Amelia II Package which has
perfectly worked for my case. However, when I started to look closely to
the values imputated I realised that the method has replaced the missing
values with negative ones which in my case was non significant and non
relevant. Beside the negative issue, I have some problem with variable
ranging between 0 and 1. Indeed, I have a variable wich representes a
percentage but when imputated, the missing values exceed the number 1 which
is wierd.
Could you have the kindness to enlighten me of a possible solution to
these two problems, knowing that by adding in the command ( ords ="X5",
p2s = 0), the negative value disappear and the imputed data for X5 are
numeric
a.out<- amelia(Mydata, m = 5, ts = "year", cs = "id", ords
="X5", p2s = 0).
This constraint (ords ="X5", p2s = 0) in the command is correct!! Thank
you in advance for your answer.
Best regards
Mouna kessentini
--
Amelia mailing list served by HUIT
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=amelia
More info about Amelia:
http://gking.harvard.edu/amelia
Amelia mailing list
Amelia(a)lists.gking.harvard.edu
To unsubscribe from this list or get other information:
https://lists.gking.harvard.edu/mailman/listinfo/amelia