20 Feb
2008
20 Feb
'08
9:23 a.m.
Hello
The manual says about identification variables: "idvars : a vector of
column numbers or column names that indicates identification variables.
These will be dropped from the analysis but copied into the imputed
datasets."
The "dropping" is not 100% clear to me in an important aspect, namely,
whether idvars are -considered- in the MI calculation process at all. So
is a) or b) the correct answer?
a) The idvars are -used- in the MI process to calculate imputations, but
are themselves not modified (imputed). They are copied to the output
dataset 1:1.
b) The idvars are only copied and -not- (even) considered in the MI
process to calculate imputations.
Thank you,
Marcus
--
Marcus M. Dapp | PhD student | ETH Zurich | www.ib.ethz.ch/people/mdapp
Prof. Thomas Bernauer, International Relations | www.ib.ethz.ch
On the shoulders of giants? http://science.creativecommons.org
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
20 Feb
20 Feb
9:27 a.m.
Hello Marcus,
The identification vars are removed from the dataset before the
algorithm is run and returned to their rightful columns once the
algorithm has finished. Thus, (b) is the correct answer. You would get
the same results from Amelia if you sent it your data with those ID
columns removed.
good luck,
matt.
On Feb 20, 2008 9:23 AM, Marcus M. Dapp <mdapp(a)ethz.ch> wrote:
Hello
The manual says about identification variables: "idvars : a vector of
column numbers or column names that indicates identification variables.
These will be dropped from the analysis but copied into the imputed
datasets."
The "dropping" is not 100% clear to me in an important aspect, namely,
whether idvars are -considered- in the MI calculation process at all. So
is a) or b) the correct answer?
a) The idvars are -used- in the MI process to calculate imputations, but
are themselves not modified (imputed). They are copied to the output
dataset 1:1.
b) The idvars are only copied and -not- (even) considered in the MI
process to calculate imputations.
Thank you,
Marcus
--
Marcus M. Dapp | PhD student | ETH Zurich | www.ib.ethz.ch/people/mdapp
Prof. Thomas Bernauer, International Relations | www.ib.ethz.ch
On the shoulders of giants? http://science.creativecommons.org
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
9:45 a.m.
New subject: [SOLVED] Are identification variables used in the MI calculations or not?
Wow, that was fast. Thank you, Matt!
Alright, that tells me that "idvars" may be used for, e.g., IDs, contact
info (like email) in survey data and so on... stuff that has no
statistical relevance whatsoever.
But, on the other side, I would not put "perfect data columns" (i.e. no
NAs) as idvars, because they would just deteriorate the MI process.
That also means that ALL columns (except idvars) in a data frame ARE
used by Amelia, either to be imputed or to be utilized for imputation
elsewhere.
I am pointing that out because I am very new to Amelia/MI -- and am
already preparing another question related to the ords and noms
parameters and related error messages I am getting. :-)
Regards,
Marcus
Am 20.02.2008 15:27 schrieb Matt Blackwell:
Hello Marcus,
The identification vars are removed from the dataset before the
algorithm is run and returned to their rightful columns once the
algorithm has finished. Thus, (b) is the correct answer. You would get
the same results from Amelia if you sent it your data with those ID
columns removed.
good luck,
matt.
On Feb 20, 2008 9:23 AM, Marcus M. Dapp <mdapp(a)ethz.ch> wrote:
> Hello
>
> The manual says about identification variables: "idvars : a vector of
> column numbers or column names that indicates identification variables.
> These will be dropped from the analysis but copied into the imputed
> datasets."
>
> The "dropping" is not 100% clear to me in an important aspect, namely,
> whether idvars are -considered- in the MI calculation process at all. So
> is a) or b) the correct answer?
>
> a) The idvars are -used- in the MI process to calculate imputations, but
> are themselves not modified (imputed). They are copied to the output
> dataset 1:1.
>
> b) The idvars are only copied and -not- (even) considered in the MI
> process to calculate imputations.
>
> Thank you,
> Marcus
>
> --
>
> Marcus M. Dapp | PhD student | ETH Zurich | www.ib.ethz.ch/people/mdapp
> Prof. Thomas Bernauer, International Relations | www.ib.ethz.ch
>
> On the shoulders of giants? http://science.creativecommons.org
> -
> Amelia mailing list served by Harvard-MIT Data Center
> [Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
>
>
--
Marcus M. Dapp | PhD student | ETH Zurich | www.ib.ethz.ch/people/mdapp
Prof. Thomas Bernauer, International Relations | www.ib.ethz.ch
On the shoulders of giants? http://science.creativecommons.org
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
11:29 a.m.
New subject: [SOLVED] Are identification variables used in the MI calculations or not?
Hi Marcus,
You are right that you want to put those variables that have no
statistical information in them in. You do want to include (i.e. not
set as "idvars") variables that are fully observed and might contain
information about the missing data. Amelia will use this information
to make better imputations.
fire away with the ordinal/nominal error messages.
regards,
matt.
On Wed, Feb 20, 2008 at 9:45 AM, Marcus M. Dapp <mdapp(a)ethz.ch> wrote:
Wow, that was fast. Thank you, Matt!
Alright, that tells me that "idvars" may be used for, e.g., IDs, contact
info (like email) in survey data and so on... stuff that has no
statistical relevance whatsoever.
But, on the other side, I would not put "perfect data columns" (i.e. no
NAs) as idvars, because they would just deteriorate the MI process.
That also means that ALL columns (except idvars) in a data frame ARE
used by Amelia, either to be imputed or to be utilized for imputation
elsewhere.
I am pointing that out because I am very new to Amelia/MI -- and am
already preparing another question related to the ords and noms
parameters and related error messages I am getting. :-)
Regards,
Marcus
Am 20.02.2008 15:27 schrieb Matt Blackwell:
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
Hello Marcus,
The identification vars are removed from the dataset before the
algorithm is run and returned to their rightful columns once the
algorithm has finished. Thus, (b) is the correct answer. You would get
the same results from Amelia if you sent it your data with those ID
columns removed.
good luck,
matt.
On Feb 20, 2008 9:23 AM, Marcus M. Dapp <mdapp(a)ethz.ch> wrote:
> Hello
>
> The manual says about identification variables: "idvars : a vector of
> column numbers or column names that indicates identification variables.
> These will be dropped from the analysis but copied into the imputed
> datasets."
>
> The "dropping" is not 100% clear to me in an important aspect, namely,
> whether idvars are -considered- in the MI calculation process at all. So
> is a) or b) the correct answer?
>
> a) The idvars are -used- in the MI process to calculate imputations, but
> are themselves not modified (imputed). They are copied to the output
> dataset 1:1.
>
> b) The idvars are only copied and -not- (even) considered in the MI
> process to calculate imputations.
>
> Thank you,
> Marcus
>
> --
>
> Marcus M. Dapp | PhD student | ETH Zurich | www.ib.ethz.ch/people/mdapp
> Prof. Thomas Bernauer, International Relations | www.ib.ethz.ch
>
> On the shoulders of giants? http://science.creativecommons.org
> -
> Amelia mailing list served by Harvard-MIT Data Center
> [Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
>
>
--
Marcus M. Dapp | PhD student | ETH Zurich | www.ib.ethz.ch/people/mdapp
Prof. Thomas Bernauer, International Relations | www.ib.ethz.ch
On the shoulders of giants? http://science.creativecommons.org
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
5935
days inactive
5935
days old
amelia@lists.gking.harvard.edu
3 comments
2 participants
participants (2)
-
Marcus M. Dapp -
Matt Blackwell