r(2000) no observations using foreach and forvalue

[Allen Smith]

Hi,

I try to run a tobit regression on the panel dataset through the following code. However, it does not successfully skip error of

 insufficient observations r(2000);

Can you give me some suggestions? I have attached a sample of my data for testing the code. Many thanks.

local depent  pat cite exi6 exr6
 
foreach dep of local depent{
     forvalues i= 1(1)10{

         tobit head1`dep'  time_treat PPETA LEV CAPEXTA Q LN_AGE SHORT_LIQ ROA RDTA  i.year i.firm_id if industry==`i' & diff!=0 & 1990<=year<=2014  ,ll(0) vce(cluster year)
        
        if c(rc)==0{
        eststo ols_`dep'_`i'_con1
        }
       else if inlist(c(rc), 2000, 2001)  {  
        //  NO OBSERVATIONS OR INSUFFICIENT OBSERVATIONS
        //  ADVISE USER OF THE SITUATIOIN, BUT AS THIS IS
        //  AN EXPECTED PROBLEM, MOVE ON
            display as error "industry=='i' : no or insufficient observations"
            
           }
       else {  
        display as error "Unexpected error when industry == `i' "
        error c(rc)
        }
    }
}

[Nick Cox]

Your code throws you out before the if ... else ... commands are seen, let alone executed. Stata does not hold on to see what is coming next.

Here is a simplified if silly analogue

. sysuse auto, clear
(1978 automobile data)

. regress mpg weight if 7 > 42
no observations
r(2000);

. noisily capture regress mpg weight if 7 > 42

. if _rc == 2000 di "no observations"
no observations

Sometimes noisily capture is what you need and sometimes capture noisily and sometimes capture.

Note: You're asked not to post .dta attachments. https://www.statalist.org/forums/help#stata