st: xtabond2-should we keep or drop the missing observations?

[ATANASOVSKA Viktorija <V.Atanasovska@staffs.ac.uk>]

Hello,

In my empirical study I use dynamic panel, i.e. xtabond2.

My sample has around 1312 observations in total, but one variable (x) has 120 missing observations.

First, I have estimated the model with the dataset as given (including the missing observations) and in the printout it is shown that 1068 obs. are used.

Second, I drop all the missing observations for X (120 obs are dropped) and after estimation 1038 obs. are used according to the printout.

I have two questions:
 1. Why my results differ with and without missing obs being included? Is there any procedure to be followed, for example missing should be excluded or not?
 2. Why the difference between the "whole" data set and the "cleaned" data set is lower than 120 (the number of obs. dropped)?

I use the orthogonal option given that my dataset is with gaps.

Thank you very much for any help regarding this issue.

Kind regards,

Viktorija Atanasovska

The information in this email is confidential and is intended solely for the addressee.  Access to this email by anyone else is unauthorised.

If you are not the intended recipient, any disclosure, copying, distribution or any action taken or omitted to be taken in reliance on it, except for the purpose of delivery to the addressee, is prohibited and may be unlawful.  Kindly notify the sender and delete the message and any attachment from your computer.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/