use a matrix to save results of looping

[Imed Limam]

Dear Statalisters,
I am trying to use a matrix to save results of looping over countries (panel data) to obtain lag-order selection statistics for VARs and VECMs and feed in the optimal lag into the matrix. However, I keep receiving this error message after running the procedure for the first country:
“new variables cannot be uniquely named or already defined, after running”

I am not able to figure out what is wrong after several attempts. Here is the part of code I am using to loop over countries:

* Identify the unique countries in your dataset
levelsof country, local(country_list)

* Initialize an empty matrix to store results
local ncount : word count `country_list'
gen Country = "`country_list'"
matrix P = J(`ncount',1,.)


local i = 1
* Loop over each country
foreach country of local country_list {
    preserve
    * Filter data for the current country
    keep if country == "`country'"
    di "Processing country:`country'"
    varsoc rI rC rY  
    matrix A = r(stats)
    svmat A, name(col)
egen minh = min(H)
gen optimal_lag = lag if minh == HQ
sum optimal_lag if optimal_lag<.
local olag = r(mean) 
matrix P[`i',1] = `olag'
matrix drop A
drop lag LL LR df p FPE AIC HQIC SBIC minh optimal_lag

   local ++i
}
matrix list P, format(%2.0f)

Any help will be kindly appreciated.

Imed