Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Мещеряков Антон <anton.mesh777@yandex.ru> |
To | statalist@hsphsun2.harvard.edu |
Subject | st:transition from unemployment |
Date | Sun, 22 Apr 2012 13:27:09 +0300 |
Thanks for your help Katya and Nick. gen month=(p2-600)*(year==2006)+(p2-700)*(year==2007) bysort id_res: egen emp1=min(month/(steoold==1)) if year==2007 bysort id_res: egen unemp2=min(month/(steoold==2)) if year==2007 bysort id_res: egen olf3=min(month/(steoold==3)) if year==2007 gen emp=1 if emp1==month replace emp=0 if emp==. gen unemp=1 if unemp2==month replace unemp=0 if unemp==. gen olf=1 if olf3==month replace olf=0 if olf==. A this moment I have three generate variables, which show the exit of unemployment at the first questioning in 2007: - emp (1-emp, 0 -othres ) - unemp (1- unemp, 0 -others) - olf (1-olf, 0 -others) Now I would like to do next: 1. identify the structure of each group, for example (in per cent): 1a. how many people who become empl, olf or unempl are females and males It was easy to do tab emp stat,r tab unemp stat,r BUT i do not know how to show this: 1.b. i need per cent of respondents who change status in emp, olf and unemp amid females and males. It may look like this tab emp stat if year==2007,col BUT i think it is wrong cause in this way it covers the whole respondents in each category during 2007 for example, employed: | Sex emp | female male | Total -----------+----------------------+---------- 0 | 314.57674 341.4202 | 655.99694 | 87.13 84.63 | 85.81 -----------+----------------------+---------- 1 | 46.453121 61.995632 | 108.44875 | 12.87 15.37 | 14.19 -----------+----------------------+---------- Total | 361.02986 403.41584 | 764.4457 | 100.00 100.00 | 100.00 Do you have any suggestions? Best regards. * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/