The name is absent

Stata Technical Bulletin

STB-22

In this example, censor generates two variables, jobl and Tjobl. jobl is the 0/1 variable that records the individual’s
change in status. Tjobl records the time—age in this example—at which the change occurs. For individual ‘1’, the transition
took place at the end of the first observation, when he passed from the not working state to the self-employed state. This
individual was 22 years old at the end of the first observation.

Censoring can occur for reasons unrelated to the phenomenon under study. In this example data set, some individuals are
lost from the survey. The variable lost is coded as ‘0’ when an individual is still in the survey and as ‘1’ when the individual
is “lost”.

The before () option of censor provides a convenient way to handle the lost individuals:

. drop jobl Tjobl

. censor id age = activity==l ∣ activity==2, generate(jobl) before(Iost==I)

. list id activity age lost jobl Tjobl

	id	activity	age	lost	jobl	TjObl
1.	1	nonactiv	22	in	1	22
2.	1	selfemp	27	in
3.	1	selfemp	39	in
4.	2	nonactiv	19	in	O	19
ε.	3	nonactiv	19	in	1	19
6.	3	selfemp	22	in	.	.
7.	3	waged	28	in	.	.
S.	4	nonactiv	7	in	O	7
9.	4	nonactiv	20	out	.	.
10.	4	selfemp	23	out	.	.
11.	ε	nonactiv	4	in	1	4
12.	ε	waged	11	in	.	.
13.	ε	nonactiv	22	in	.	.
14.	ε	waged	28	in	.	.
ιε.	6	nonactiv	12	in	O	12
16.	6	nonactiv	24	out	.	.
17.	6	waged	27	out	.	.
18.	7	nonactiv	24	in	O	24
19.	8	nonactiv	16	in	O	16
20.	8	nonactiv	39	out

Suppose we need to create two censoring variables for two competing risks; for example, wage-work and self-employment.
censor can handle this case as well:

. censor id age = activity==!, generate(waged) before(Iost==I ∣ activity==2)

. censor id age = activity==2, generate(self) before(lost==l ∣ activity==l)

. list id activity age waged Twaged self Tself, nodisplay

	id	activity	age	waged	Twaged	self	Tself
1.	1	nonactiV	22	O	22	1	22
2.	1	selfemp	27
3.	1	selfemp	39
4.	2	nonactiV	19	O	19	O	19
ε.	3	nonactiV	19	O	19	1	19
6.	3	selfemp	22	.	.	.	.
7.	3	waged	28	.	.	.	.
8.	4	nonactiV	7	O	7	O	7
9.	4	nonactiV	20	.	.	.	.
10.	4	selfemp	23	.	.	.	.
11.	ε	nonactiV	4	1	4	O	4
12.	ε	waged	11	.	.	.	.
13.	ε	nonactiV	22	.	.	.	.
14.	ε	waged	26	.	.	.	.
ιε.	6	nonactiV	12	O	12	O	12
16.	6	nonactiV	24	.	.	.	.
17.	6	waged	27	.	.	.	.
18.	7	nonactiV	24	O	24	O	24
19.	8	nonactiV	16	O	16	O	16
20.	8	nonactiV	39

More intriguing information

1. The name is absent
2. Cultural Neuroeconomics of Intertemporal Choice
3. The name is absent
4. Partner Selection Criteria in Strategic Alliances When to Ally with Weak Partners
5. AGRICULTURAL TRADE LIBERALIZATION UNDER NAFTA: REPORTING ON THE REPORT CARD
6. The name is absent
7. The Institutional Determinants of Bilateral Trade Patterns
8. Økonomisk teorihistorie - Overflødig information eller brugbar ballast?
9. ENERGY-RELATED INPUT DEMAND BY CROP PRODUCERS
10. Agricultural Policy as a Social Engineering Tool