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" <statalist@hsphsun2.harvard.edu> |

Subject |
Re: st: xtile creating different deciles using same data |

Date |
Tue, 3 Jan 2012 12:59:40 +0000 |

Nick

Just checked, and it isn´t. best, Alvaro. On 2 January 2012 22:01, Nick Cox <njcoxstata@gmail.com> wrote:Back to the Stata question: Is -weight- also constant within households? NickOn Mon, Jan 2, 2012 at 8:23 PM, Cameron McIntosh<cnm100@hotmail.com> wrote:A comment on the nature of your income variable, as this is not atrivial matter. Per capita income is indeed preferable to rawtotal household income but is still not optimal, for reasonsdiscussed in:Carson, J. (2002). Family spending power. Perspectives on Labourand Income, 3(10), 24-32.http://www.statcan.gc.ca/studies-etudes/75-001/archive/e-pdf/5018698-eng.pdfI would suggest weighting in the manner Carson suggests (orslightly differently if the context warrants), and see if this hasan impact on the results. Perhaps some of the OECD measures mightbe useful as well:http://www.oecd.org/document/51/0,3746,en_2649_33933_49147827_1_1_1_1,00.htmlFrom: A.N.Herrera@sussex.ac.ukHi, I am using the command xtile on stata 11, 32bits, to createincomedeciles on my database, but I found an inconsistency:I have a variable with the household per capita income, with datafor thewhole population. Then I create deciles, but I don't do it over the population, but households.To do so, I use only one observation per household (they allshare thesame household per capita income) to create my deciles, and thenI assignthe rest of the household members to the decile of suchobservations.Of course, as poorer families tend to be larger, I end up withdeciles thathave more than 10% of the population on the lower end of thedistribution,and others with less than 10% on the other end. That's fine withme.basically, what I do is xtile decaux==income if count==1 [w=weight], nq(10) wherecount==1 is the first member-chosen randomly- of each household,andthen recode decaux .=0 by id_househ: egen decile=sum(decaux) I assign the other members of each household to the deciles of their respective members (count==1)The problem is that if I run the same commands on the samedatabase for asecond time (or a third, or fourth, always without modifying thedata),then the number of observations assigned to each decile differsevery time.The overall population does not change, but the populationassigned to thedeciles changes marginally every time.

* * 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/

**Follow-Ups**:**Re: st: xtile creating different deciles using same data***From:*johnataguba@yahoo.co.uk

**References**:**st: xtile creating different deciles using same data***From:*"Alvaro Herrera E." <A.N.Herrera@sussex.ac.uk>

**RE: st: xtile creating different deciles using same data***From:*Cameron McIntosh <cnm100@hotmail.com>

**Re: st: xtile creating different deciles using same data***From:*Nick Cox <njcoxstata@gmail.com>

**Re: st: xtile creating different deciles using same data***From:*"Alvaro Herrera E." <A.N.Herrera@sussex.ac.uk>

- Prev by Date:
**Re: st: standard errors after xtmixed, predit.., fitted** - Next by Date:
**st: Difference in values within the same variable** - Previous by thread:
**Re: st: xtile creating different deciles using same data** - Next by thread:
**Re: st: xtile creating different deciles using same data** - Index(es):