How to create a new binary variable (0-1) based on some conditions in Stata?

[smith Jason]

input id t1 t2 t3
1 1.5 1.4 1.6
2 1.5 . 1.9
3 . 2.0 1.6
4 . . 2.2
5 1.9 2.1 2
6 1.8 2.0 1.9
7 . . .
end


As suggested above, I want to create a new variable called sign.
If t2>t1, then sign=1; if t2<=t1, then sign=0.
Because t1 and t2 all have missing values, how can I keep missing values in the new variable "sign" in Stata?
I think the rule is as long as there is a missing value in t1 or t2, then the variable "sign" should be missing value.
Thanks for your Stata code!

[smith Jason]

up!

[Joro Kolev]

This will probably do it, but do check in your full dataset:

. gen sign = .
(7 missing values generated)

. replace sign = 0 if t2<=t1 & !missing( t1)
(1 real change made)


. replace sign = 1 if t2>t1 & !missing( t2)
(2 real changes made)

[smith Jason]

Thank you!

[Nick Cox]

Bumping a post after 26 minutes is not good practice -- if that is what #2 is. We explain that already at #1 of https://www.statalist.org/forums/help#adviceextras

gen wanted = t2 > t1 if !missing(t1, t2)

is another way to do it with less code.

[smith Jason]

Thank you very much!