[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Ekaterina Hertog <ekaterina.hertog@sociology.ox.ac.uk> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: AW: How to generate a variable equalling an earliest date of action b y individual? |

Date |
Mon, 21 Sep 2009 21:34:43 +0100 |

Dear Martin, Thank you very much for the advice! This last solution definitely produces stable results so solves the problem of outcome variation. I am still designing a way to ensure the code deals correctly with my numerous instances of tied data but so far it seems to work. Thanks a lot again! warm regards, Ekaterina In message <000b01ca3947$39d063a0$ad712ae0$@weiss1@gmx.de> statalist@hsphsun2.harvard.edu writes: > > <> > > " Out of this dataset I want to keep for each individual the date he or she > made his first approach and the date he or she was first accepted." > > > Ekaterina might also want to try -collapse- for this problem setup: > > > ************* > clear* > > inp byte id /* > */ str20(dateapproached dateaccepted ) > 1 "20.04.2007" " " > 1 "20.04.2007" " " > 1 "18.05.2007" "20.05.2007" > 1 "18.05.2007" " " > 2 "19.01.2006" " " > 2 "20.03.2007" "24.03.2007" > 2 "20.03.2007" " . " > 3 "11.09.2007" "18.09.2007" > 3 "11.09.2007" "15.09.2007" > 3 "01.10.2007" " " > 3 "07.11.2007" " " > 4 "06.06.2007" " " > 4 "08.07.2007" " " > end > > replace dateaccepted= /* > */ trim(dateaccepted) > > gen dateapproached2= /* > */ date(dateapproached, "DMY") > format dateapproached2 /* > */ %tdMonth_DD,_CCYY > > gen dateaccepted2= /* > */ date(dateaccepted, "DMY") > format dateaccepted2 /* > */ %tdMonth_DD,_CCYY > > drop dateapproached /* > */ dateaccepted > > compress > list, noobs sepby(id) > > collapse (first) dateapproached2 /* > */ (firstnm) dateaccepted2, /* > */ by(id) > > list, noobs > ************* > > > > HTH > Martin > > > -----Ursprüngliche Nachricht----- > Von: owner-statalist@hsphsun2.harvard.edu > [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Ekaterina > Hertog > Gesendet: Samstag, 19. September 2009 18:23 > An: statalist@hsphsun2.harvard.edu > Betreff: st: How to generate a variable equalling an earliest date of action > b y individual? > > Dear all, > I have a dataset of individuals each of whom approaches a number of others > and after approaching each person is either accepted or rejected. I would > like to do an event history analysis of how the duration of time it takes > for an individual to secure an acceptance depends on his or hers individual > characteristics. > The variables I am interested in have the following structure (missing > dateacceppted means a rejection): > > id dateapproached dateaccepted > 1 20.04.2007 . > 1 20.04.2007 . > 1 18.05.2007 20.05.2007 > 1 18.05.2007 . > 2 19.01.2006 . > 2 20.03.2007 24.03.2007 > 2 20.03.2007 . > 3 11.09.2007 18.09.2007 > 3 11.09.2007 15.09.2007 > 3 01.10.2007 . > 3 07.11.2007 . > 4 06.06.2007 . > 4 08.07.2007 . > > Out of this dataset I want to keep for each individual the date he or she > made his first approach and the date he or she was first accepted. > To do so I tried to create a new variable date1accept which for each > individual has the same value => the date he or she was first accepted. > After that I dropped all the observations, but the 1st one for each > individual. > To do so I used the following code: > sort id dateaccept > foreach X of new date1accept { > by id : gen `X' = dateaccept[1] > } > > sort id > by id: gen y=_n > keep if y==1 > > I then ran two tests and Stata confirmed that thes assertions below were > true: > /*To check that all the values of date1accept are the same*/ > bysort id (date1accept) : assert date1accept[1] == date1accept[_N] > /*To check that dat1accept = the lowest value of dateaccept*/ > bysort id (dateaccept) : assert dat1accept[1] == dateaccept[1] > > yet something seems to be wrong with my commands, each time I re-run the > code it comes up with slightly different set of date1accept values. I cannot > understand where can my code be wrong so my tests do not pick it up. > Could anyone suggest what might be wrong or offer an alternative solution > for my problem? > I will be most grateful for any advice, > Warm regards, > Ekaterina > > * > * 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/ > > > * > * 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/ -- Ekaterina Hertog (nee Korobtseva) Nissan Institute of Japanese Studies 27 Winchester Road, Oxford OX2 6NA * * 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/

**References**:**st: outreg2 and xtmixed?***From:*Christian Weiß <mail@cweiss.org>

**RE: st: outreg2 and xtmixed?***From:*Roy Wada <roywada@hotmail.com>

**Re: st: outreg2 and xtmixed?***From:*Christian Weiß <mail@cweiss.org>

**RE: st: outreg2 and xtmixed?***From:*"Martin Weiss" <martin.weiss1@gmx.de>

**st: How to generate a variable equalling an earliest date of action b y individual?***From:*Ekaterina Hertog <ekaterina.hertog@sociology.ox.ac.uk>

**st: AW: How to generate a variable equalling an earliest date of action b y individual?***From:*"Martin Weiss" <martin.weiss1@gmx.de>

- Prev by Date:
**AW: st: AW: why is this graph all over the place?** - Next by Date:
**Re: st: AW: why is this graph all over the place?** - Previous by thread:
**st: AW: How to generate a variable equalling an earliest date of action b y individual?** - Next by thread:
**st: multiple imputation** - Index(es):

© Copyright 1996–2016 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |