Dear all,
I’m encountering some troubles with the ivreg2 command / package.
To be more precise, I’m facing the error message “Mata run time error r(3351)”, when I’m running an IV regression of a subsample of my regression sample. Command: ivreg2 vary (varx1=varz) varx2 varx3 ... if regsample == 1, cluster(hh) ffirst savefirst savefprefix(first1)
In fact, I run regressions for a German household dataset at all and afterwards I’m separating this dataset into East- and West German observations. (The number of observations for West Germany is roughly twice as high as the number of observations for East Germany.) Whereas it works fine for West German observations and the whole sample, this above-mentioned error message occurs for the subsample of East German observations.
In total, this problem strongly resembles the following topic --> https://www.statalist.org/forums/for...run-time-error . However, following the given solutions to this earlier topic and searching for other explanations with Google didn’t help me.
Concerning my environment: I’m using Stata SE 13.1 on a Linux Server in a high data protection environment. Therefore, I have no admin rights and I am unfortunately not able to directly test any solutions. (Sorry for this inconvenience.) The package ivreg2 is of version 4.1.10 9Feb2016 and was reinstalled by an admin two days ago. (ranktest is also installed and up to date).
I can rule out that there any missings in the observations I used leading to this error message, since I am building a regressions sample beforehand with the e(sample) function.
Does anyone have a solution for this problem?
Thanks in advance!
Best,
Andreas
I’m encountering some troubles with the ivreg2 command / package.
To be more precise, I’m facing the error message “Mata run time error r(3351)”, when I’m running an IV regression of a subsample of my regression sample. Command: ivreg2 vary (varx1=varz) varx2 varx3 ... if regsample == 1, cluster(hh) ffirst savefirst savefprefix(first1)
In fact, I run regressions for a German household dataset at all and afterwards I’m separating this dataset into East- and West German observations. (The number of observations for West Germany is roughly twice as high as the number of observations for East Germany.) Whereas it works fine for West German observations and the whole sample, this above-mentioned error message occurs for the subsample of East German observations.
In total, this problem strongly resembles the following topic --> https://www.statalist.org/forums/for...run-time-error . However, following the given solutions to this earlier topic and searching for other explanations with Google didn’t help me.
Concerning my environment: I’m using Stata SE 13.1 on a Linux Server in a high data protection environment. Therefore, I have no admin rights and I am unfortunately not able to directly test any solutions. (Sorry for this inconvenience.) The package ivreg2 is of version 4.1.10 9Feb2016 and was reinstalled by an admin two days ago. (ranktest is also installed and up to date).
I can rule out that there any missings in the observations I used leading to this error message, since I am building a regressions sample beforehand with the e(sample) function.
Does anyone have a solution for this problem?
Thanks in advance!
Best,
Andreas
