Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

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

From |
Nick Cox <njcoxstata@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: Sorting relationship dates to get a correct chronological sequence |

Date |
Mon, 10 Jan 2011 10:17:39 +0000 |

The extended missings .a ... .z may be of use. However, they can't easily be combined with real dates as 2011 < .a regardless. Nick On Mon, Jan 10, 2011 at 10:05 AM, Anna Reimondos <areimondos@gmail.com> wrote: > Dear Nick > Thanks for the reply, I see what you mean. It looks like I will need > to do some work on the data to achieve my desired result; I will try > imputing the dates based on some logical rules about the relationship > orders. > > Cheers > Anna > > On Mon, Jan 10, 2011 at 8:31 PM, Nick Cox <njcoxstata@gmail.com> wrote: >> Your problem does not look soluble to me unless you impute dates that >> are consistent with what you can infer about the time order of >> relationships. >> >> Nick >> >> On Mon, Jan 10, 2011 at 2:50 AM, Anna Reimondos <areimondos@gmail.com> wrote: >> >>> I am having trouble thinking of a way to sort my data they way I would like to. >>> >>> What I am trying to do is to sort people's relationships in the >>> chronological order that they happened. >>> >>> I have separate pieces of information about the start and end dates of >>> all 1) cohabitations 2) marriages. >>> >>> EXAMPLE >>> A person might have had 5 relationships: 3 cohabitations and 2 >>> marriages. The second marriage is still ongoing. >>> In long format the data looks like: >>> >>> Person Relationship Star Year End Year >>> 1 Cohabitation 1 2001 2001 >>> 1 Cohabitation 2 -1 missing -1 missing >>> 1 Cohabitation 3 -1 missing -1 missing >>> 1 Marriage 1 1998 2000 >>> 1 Marriage 2 2008 -20 (Ongoing) >>> >>> >>> I have tried sorting the data by start year, to get the chronological >>> order correct but obviously run into problems with the relationships >>> that have missing start or end dates since these have negative values >>> and therefore appear first, as such: >>> >>> Person Relationship Star Year EndYear >>> 1 Cohabitation 2 -1 missing -1 missing >>> 1 Cohabitation 3 -1 missing -1 missing >>> 1 Marriage 1 1998 2000 >>> 1 Cohabitation 1 2001 2001 >>> 1 Marriage 2 2008 -20 (Ongoing) >>> >>> >>> However, this is wrong because obviously Cohabitation 2 and >>> Cohabitation 3 must have occured *after* Cohabitation 1. >>> The actual correct order I am trying to achieve is : Marriage 1, >>> Cohabitation 1, Cohabitation 2, Cohabitation 3, Marriage 2. >>> I was hoping someone would have a suggestions of how I could achieve >>> the correct sequence. I think I will have to use a loop maybe? * * 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: Sorting relationship dates to get a correct chronological sequence***From:*Anna Reimondos <areimondos@gmail.com>

**Re: st: Sorting relationship dates to get a correct chronological sequence***From:*Nick Cox <njcoxstata@gmail.com>

**Re: st: Sorting relationship dates to get a correct chronological sequence***From:*Anna Reimondos <areimondos@gmail.com>

- Prev by Date:
**Re: st: Sorting relationship dates to get a correct chronological sequence** - Next by Date:
**Re: st: question about the Roy zelner test of poolability** - Previous by thread:
**Re: st: Sorting relationship dates to get a correct chronological sequence** - Next by thread:
**st: Quick doubt about the 2-way clustering option used in IV_GMM2s** - Index(es):