Friday, February 17, 2006
Dear Statalisters: I have recently passed to xtabond2 in order to estimate a dynamic panel using the GMM in Difference, the GMM in Levels and the System GMM estimators. Although, I have the updated version of xtabond2 (2.1.2, 8Jan2006) I am struggling to estimate my model. To exploit the speed improvements of the new xtabond2 I set matsize 2000 and set matafavor speed. Moreover, to minimize the size of my data I keep only the variables that I need. However, after 15' of waiting I had to click on break and try to estimate the model favoring space over speed. Again I had to wait for a long time without getting any output. My last choice was to estimate the model using the nomata option. Surprisingly, I got my results within 2'. When I tried to regress the model using another computer I got the following reading:
xtabond2_mata(): 3900 unable to allocate real <tmp>[45984,843] <istmt>: - function returned error. Another statalister had the same problem in the past. Is there any solution to this issue? Is it better to stick with the nomata option?
My second question is regarding the equation() suboption. I am quite puzzled about the syntax of this option The following is the syntax of my estimations:
i) First-Difference: xtabond2 depvar l.depvar l(0/1). (endogvars) exogvars, gmm(depvar endogvars, lag(2 3)) iv(exogvars) noleveleq robust small twostep
Do I have to specify equ(diff) in the gmmstyle and the ivstyle?
ii)Level: xtabond2 depvar l.depvar l(0/1). (endogvars) exogvars, gmm(depvar endogvars, lag(2 3)equ(lev)) iv(exogvars,equ(lev)) robust small twostep
iii) System: xtabond2 depvar l.depvar l(0/1). (endogvars) exogvars, gmm(depvar endogvars, lag(2 3)) iv(exogvars) robust small twostep
Do I have to type equ(both) in the gmmstyle and the ivstyle?
Any advices and comments would be appreciated.
This message has been checked for viruses but the contents of an attachment may still contain software viruses, which could damage your computer system: you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
* * For searches and help try: * http://www.stata.com/support/faqs/res/findit.html * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/
Links to this post: