I am using a cleaned, balanced panel on retirement behaviour with Stata 16. One of the questions I'm attempting to address is the impact of retirement on self-reported health.
I am using a regression discontinuity design, with -
  1. Outcome variable: selfhealth (Self reported health index)
  2. Running variable: z_age (Age in years - 66.5, retirement age)
  3. Treatment variable: retire (1 if the individual is not in the labour force)
Here is an example of the data I have -


Code:
* Example generated by -dataex-. To install: ssc install dataex
clear
input long hhidpn byte(wave selfhealth) float(z_age retire)
1010  1 4 -12.5 0
1010  2 4 -10.5 0
1010  3 .     . .
1010  4 .     . .
1010  5 .     . .
1010  6 .     . .
1010  7 .     . .
1010  8 .     . .
1010  9 .     . .
1010 10 .     . .
1010 11 .     . .
1010 12 .     . .
1010 13 .     . .
2010  1 3  -9.5 1
2010  2 3  -7.5 0
2010  3 4  -4.5 1
2010  4 5  -3.5 1
2010  5 3  -1.5 1
2010  6 .     . .
2010  7 .     . .
2010  8 .     . .
2010  9 .     . .
2010 10 .     . .
2010 11 .     . .
2010 12 .     . .
2010 13 .     . .
3010  1 4 -10.5 0
3010  2 4  -8.5 0
3010  3 3  -6.5 1
3010  4 3  -4.5 1
3010  5 4  -2.5 1
3010  6 3   -.5 1
3010  7 3   1.5 1
3010  8 3   3.5 1
3010  9 3   5.5 1
3010 10 3   7.5 1
3010 11 3   9.5 1
3010 12 .     . .
3010 13 .     . .
3020  1 3 -12.5 1
3020  2 3 -11.5 1
3020  3 3  -9.5 1
3020  4 3  -7.5 1
3020  5 3  -5.5 1
3020  6 3  -3.5 1
3020  7 3  -1.5 1
3020  8 3    .5 1
3020  9 4   2.5 1
3020 10 4   5.5 1
3020 11 4   6.5 1
end
label values selfhealth SHLT
label def SHLT 3 "3.good", modify
label def SHLT 4 "4.fair", modify
label def SHLT 5 "5.poor", modify
The command I'm running is
Code:
rdrobust selfhealth z_age, fuzzy(retire) vce(cluster hhidpn)
I get an error with the following output while using this-
Code:
. rdrobust selfhealth z_age, fuzzy(retire) vce(cluster hhidpn)
                 <istmt>:  3301  subscript invalid
r(3301);
Like a few previous posts on this error mentioned, I made sure to update Stata and the rdrobust package. However, I'm still running into this problem. I'd appreciate any help on this!