Question about writing a programme that takes on different values based on the percentile

[Goh jedrek]

I am supposed to write a programme called vx, that takes on the value of 1 if the variable’s value is in the top 5th percentile and 0 otherwise and error if input variable is categorical variable.

capture program drop vx
program vx
syntax varname
su `varlist'
if var == factor {
display "Invalid argument: Input variable is not a categorical variable"
}
else var >= `r(p95)' {
display "1"
}
else var < `r(p95)' {
display "0"
}
}
end

I ran the programme on 3 variables: length, mass and weight. I ended up with the error 111, which said var could not be found. Is there a way to remedy the programme?

[Nick Cox]

Is this an assignment? See our guidance at FAQ Advice Extras.

The central technique you presumably need is

. sysuse auto, clear
(1978 automobile data)

. summarize price, detail

                            Price
-------------------------------------------------------------
      Percentiles      Smallest
 1%         3291           3291
 5%         3748           3299
10%         3895           3667       Obs                  74
25%         4195           3748       Sum of wgt.          74

50%       5006.5                      Mean           6165.257
                        Largest       Std. dev.      2949.496
75%         6342          13466
90%        11385          13594       Variance        8699526
95%        13466          14500       Skewness       1.653434
99%        15906          15906       Kurtosis       4.819188

. gen wanted = price >= r(p95) if price < .

. tab wanted

     wanted |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |         70       94.59       94.59
          1 |          4        5.41      100.00
------------+-----------------------------------
      Total |         74      100.00

The evident problem with your program is that Stata has no idea what you mean by var, as it's not a variable or scalar in the dataset. The code also seems confused between looking at an entire variable and looking at individual values.