Pierre

Were you referring to the model for USA with pure rate of time preference derived from real interest rates (USA_real)? I only had two missing or incomplete observations dropped in the estimation; the model is saved to the icon view within the attached session file.

For your reference, I am using gretl 1.9.2-cvs on Debian GNU Linux 5.0 (Lenny).

Yangbo Du

Pierre Chausse wrote:

Hi,

I tried it with my Mac version and it does not crash. However, your last estimation returns an error message.
You have so many missing values in your instruments and dependent variable that the second stage OLS cannot be computed. In the first stage OLS 31 observations are dropped because of missing values. In the the second stage, there are only 3 observations can coincide. I don't know why it crashed, but I know that you cannot estimate that model.

On 10-11-25 11:45 AM, Yangbo Du wrote:
Whenever I run TSLS using the attached script file for each country's pure rate of time preference across three regressors (each of the three UN HDI components) and three instruments, gretl crashes after only a few regression runs. Any assistance with resolving this issue will be highly appreciated. The data file and the output from running the script are also attached.

Yangbo Du
_______________________________________________
Gretl-users mailing list
Gretl-users@lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users
--
Pierre Chaussé
Assistant Professor
Department of Economics
University of Waterloo
_______________________________________________
Gretl-users mailing list
Gretl-users@lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users