Error after xtcsd command

[Alan Brito]

Dear users,

whenever I run the user-written xtcsd command (under its three options, pesaran, friedman, or fress), I get the same error:

_merge already defined

This command is to be run after an xtreg, re or xtreg, fe model. The error happens in a complex model with many regressors and in a trivial model with only one regressor.

Any idea?

[William Lisowski]

At some point in creating the dataset input to xtcsd, you used the merge command to combine other datasets. The merge command creates the variable _merge by default, and apparently in its work, xtcsd is also using merge and attempting to create _merge. Your solution is to drop or rename your _merge variable before using the xtcsd command.

[Alan Brito]

My database does not have any _merge variable name. Have you tried the command to see if it work?

[Zeqi Yasin]

I also have the same problem. May I know can you finally deal with this?

[Neha Kanojia]

Dear Users,

When I am using the command "xtcsd, pesaran abs"
then I got the error that the panel is highly unbalanced.

"Error: The panel is highly unbalanced. Not enough common observations across the panel to perform Pesaran's test. insufficient observations "

so, is there any other way to calculate cross-sectional independence? In my dataset N=80 and T=19 years.

Best Regards
Neha