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: Multiple imputation for longitudinal data
From
Clara Barata <[email protected]>
To
[email protected]
Subject
Re: st: Multiple imputation for longitudinal data
Date
Fri, 3 Dec 2010 14:34:01 -0500
Hi Eduardo,
A possible (albeit possibly limited) first approach would be to
arrange the data in wide rather than stacked format and impute that
way using Stata's MI. I think this is what Stas meant when he
mentioned the need to first -reshape- your data to make it one line
for each patient.
Clara
On Fri, Dec 3, 2010 at 12:20 PM, Eduardo Nunez <[email protected]> wrote:
>
> Based on what you wrote, I imagine Stata hasn't implemented these
> methods ( that utilize the
> monotonicity of monotonicity).
> Would you guide me to the software that has these estimation methods.
> Do you know if it is implemented in R?
>
>
> Best regards,
>
> Eduardo
>
>
>
> On Fri, Dec 3, 2010 at 11:50 AM, Stas Kolenikov <[email protected]> wrote:
> >
> > What I am saying is that there are estimation methods that utilize the
> > monotonicity of monotonicity. Of course you can the existing methods
> > to produce something sensible. The monotone options, however, are
> > designed to work across the data set, not along the data set, so you
> > would want to -reshape- your data to make it one line for each
> > patient.
> >
> > On Fri, Dec 3, 2010 at 10:34 AM, Eduardo Nunez <[email protected]> wrote:
> > > Thank you, Stas.
> > > I can handle the monotone missing pattern either using ICE or MI
> > > impute while specifying the monotone option.
> > > But my question goes further on how to account for clustering on
> > > patient ID (which is the cluster unit)?.
> > > Should I impute data separately for each patient? or include pteID
> > > variable in the imputation model?
> > >
> > > I appreciate any help.
> > >
> > > Eduardo
> > >
> > >
> > >
> > >
> > > On Thu, Dec 2, 2010 at 6:39 PM, Stas Kolenikov <[email protected]> wrote:
> > >> You have monotone missing data, and you would most likely be better
> > >> off utilizing the methods for monotone missing data rather than
> > >> bluntly rely on multiple imputation. Check Little and Rubin's book on
> > >> missing data, chapter 7 (in the 2nd edition).
> > >>
> > >> On Thu, Dec 2, 2010 at 5:11 PM, Eduardo Nunez <[email protected]> wrote:
> > >>> Dear Statalisters,
> > >>>
> > >>> I have Stata 11.1 (MP - Parallel Edition).
> > >>>
> > >>> I am interested in performing multiple imputation on a longitudinal
> > >>> data (on several variables with a percent of missing between 1-15%),
> > >>> were subjects are the cluster units with few observations in time.
> > >>> See below the data structure:
> > >>>
> > >>> xtdes, pattern(1000)
> > >>>
> > >>> pid: 1, 2, ..., 1438 n = 1432
> > >>> visit: 1, 2, ..., 12 T = 12
> > >>> Delta(visit) = 1 unit
> > >>> Span(visit) = 12 periods
> > >>> (pid*visit uniquely identifies each observation)
> > >>>
> > >>> Distribution of T_i: min 5% 25% 50% 75% 95% max
> > >>> 1 1 1 2 3 6 12
> > >>>
> > >>> Freq. Percent Cum. | Pattern
> > >>> ---------------------------+--------------
> > >>> 650 45.39 45.39 | 1...........
> > >>> 359 25.07 70.46 | 11..........
> > >>> 202 14.11 84.57 | 111.........
> > >>> 91 6.35 90.92 | 1111........
> > >>> 52 3.63 94.55 | 11111.......
> > >>> 44 3.07 97.63 | 111111......
> > >>> 11 0.77 98.39 | 1111111.....
> > >>> 9 0.63 99.02 | 11111111....
> > >>> 6 0.42 99.44 | 111111111...
> > >>> 4 0.28 99.72 | 1111111111..
> > >>> 3 0.21 99.93 | 11111111111.
> > >>> 1 0.07 100.00 | 111111111111
> > >>> ---------------------------+--------------
> > >>> 1432 100.00 | XXXXXXXXXXXX
> > >>>
> > >>> The article included in Stata FAQ ("How can I account for clustering
> > >>> when creating imputations with mi impute?") suggested using a
> > >>> "multivariate
> > >>> normal model to impute all clusters simultaneously" or strategy 3,
> > >>> although mentioned that is best suited to balanced repeated-measures
> > >>> data.
> > >>>
> > >>> Clearly, my data is not balanced. Moreover, the percent of data
> > >>> missing increased as patient follow-up gets far from baseline.
> > >>>
> > >>> Is there any other method suited for this type of longitudinal data?
> > >>> If not, how stringent is the limitation of not being balanced.
> > >>>
> > >>> Please, any help is welcome!
> > >>>
> > >>>
> > >>> Eduardo
> > >>> *
> > >>> * 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/
> > >>>
> > >>
> > >>
> > >>
> > >> --
> > >> Stas Kolenikov, also found at http://stas.kolenikov.name
> > >> Small print: I use this email account for mailing lists only.
> > >>
> > >> *
> > >> * 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/
> > >>
> > >
> > > *
> > > * 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/
> > >
> >
> >
> >
> > --
> > Stas Kolenikov, also found at http://stas.kolenikov.name
> > Small print: I use this email account for mailing lists only.
> >
> > *
> > * 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/
>
> *
> * 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/
*
* 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/