Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: re: xtabond2 error

From   Hussain Samad <>
To   "" <>
Subject   Re: st: re: xtabond2 error
Date   Fri, 2 Dec 2011 10:56:41 -0800 (PST)

Dear Kit:

Thanks a lot for your help. I am in fact using the lag suboption and still getting the error. Here is my command:

xtabond2 mmisspay l.mmisspay week week2 pastmisswk pastmissgrpwk ageweek-hhstructweek nmembersweek age_grpweek-hhstruct_grpweek if sex==1,  gmm(mmisspay pastmisswk pastmissgrpwk, lag(2 2)) iv(ageweek-hhstructweek nmembersweek age_grpweek-hhstruct_grpweek, eq(level)) cluster(centerid);

Thanks again. Hussain

----- Original Message -----
From: Christopher Baum <>
To: "" <>
Sent: Friday, December 2, 2011 1:15 PM
Subject: st: re: xtabond2 error

I am getting following error message while running a GMM model with xtabond2:

editmissing():  3900  out of memory
  _ParseInsts():     -  function returned error
xtabond2_mata():     -  function returned error
<istmt>:     -  function returned error

I have updated my Stata from 11.0 to 11.2 as someone suggested that it will solve the problem, but I am still getting the same error message and feel really helpless. I have over 56,000 observations with more than 400 time periods (this is a weekly panel which is unbalanced).  

As in my recent reply to another posting, with 400 time periods, you are trying to create THOUSANDS of additional instrument variables unless you restrict the GMM-style instruments with the lag( ) suboption. See the description of that suboption. You probably don't want to go back more than a few time periods.


Kit Baum   |   Boston College Economics & DIW Berlin   |
                             An Introduction to Stata Programming  |
  An Introduction to Modern Econometrics Using Stata  |

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index