Re: st: Using mifit after ice to get imputed values

[Ilke UNDER <onurunder@gmail.com>]

Maarten you are correct. My purpose was to get the estimates of the
missing values. Because, I was advised to use MI before propensity
score matching --my covariates have missing values. Apperantly, I need
to apply PSM in the imputed data sets separately and try to combine my
estimates.

Thanks for the clarification, this helps a lot!

Onur

On Sat, Jul 17, 2010 at 3:02 AM, Maarten buis <maartenbuis@yahoo.co.uk> wrote:
> --- On Sat, 17/7/10, Ilke UNDER wrote:
>> However, still cannot find a way to get the final
>> imputed values for my initial data set.
>
> It just occured to me that you might be looking for
> an estimate (one number) of the missing value. This
> would be a (commonly made) mistake. Multiple Imputation
> is not about providing estimates for the missing values,
> that information is considered to be lost forever. It
> is about making the best use of the information that is
> available from those observations that contain missing
> values. The imputation part is just a technical devise
> used for integration over the distribution of likely
> values for the missing value, it is not an estimate
> of the value of the missing value.
>
> Hope this helps,
> Maarten
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
> http://www.maartenbuis.nl
> --------------------------
>
>
>
>
> *
> *   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/