Thursday, January 05, 2006

st: Stata versions and reproductibility problem.


I have a problem. I was running an old code today and it crashed. All the graph commands are not working.

Previsously, with stata se 8.2, the program worked perfect. In the mean time, I desinstall version 8.2 for version 9 and now 9.1 to gain some space.

I went ahead and put a // in front of all my graph commands (kdensity, twoway, graph...). The second problem I discover was that compared to my previous results, stata 9.1 provides results with are differents in the decimal points (for example, 42.84% of poverty rate and now 42.12%). I did not change anything to the code except "cd" and "//".

What happens if a stata program is deeply modified (say, by updates through ssc archives or by StataCorp)? Especially for reproductibility issues?

Best regards.


PS: I have connection problems. So apologies if this mail pops up twice.

Amadou DIALLO. Poverty and Health Specialist AFTPM, Africa Region. The World Bank. (202) 458 0288.

* * For searches and help try: * * *


Links to this post:

Create a Link

<< Home

This page is powered by Blogger. Isn't yours?