Hi everyone.
I am a novice of Stata and English is not my mother language. So I will try my best to explain as clear as possible the problem that I have run into.
I am currently using Stata 15.0 and my data having 20 variables. i created every possible combination out of them by using -tuples (SSC install). After that, I regress my dependent variable to every single combination by using -forvalues
in order to capture the estimation result, coefficients, bayesian information criterion. Unfortunately, something went wrong with my loop and i constantly got the same result for every regression. Stata somehow forgot the "combination" part that i would have wanted to added in the regression equation to be exact.
Here is my coding.

*Create variable to capture coefficients estimated.
foreach var of varlist `r(varlist)'{
gen b_`var'= .
}

ds, has(type double)
tuples `r(varlist)', varlist //SSC install tuples//

*looping

forval i = 1/`ntuples' {
probit GFC `tuple`i'' i.year
est sto model`i'
foreach var in `tuple`i'' {
scalar b_`var' = _b[`var']
replace b_`var' = `=scalar(b_`var')' in `i'
}
}

Stata somehow constantly returns the result of this regression: probit GFC i.year. I have tried many ways but it did not work at all.
Could you please explain what happened and show me how to fix it.
Many thanks and best regards