[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

From |
Wolfgang Langer <langer@soziologie.uni-halle.de> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: Extracting the varname of e(wexp) after regress ? Anotherquestion more |

Date |
Mon, 14 Nov 2005 18:29:04 +0100 |

Wolfgang Langer wrote:

Dear Statalisters,n j cox wrote:This question was answered promptly:

http://www.hsph.harvard.edu/cgi-bin/lwgate/STATALIST/archives/statalist.0511/date/article-336.html

>>> Wolfgang Langer wrote

Sorry, I try it a second time because I haven't got an answer.

I have written a small program to calculate the

goodness-of-fit-chi2-test after estimating a WLS-regression modell for

tabulated data (GSK-approach). How can I extract the varname from the

e(wexp) container to use it in my program? Typing "dis e(wexp)" I get

only the expression "=varname of weighting variable" but not the

varname itself. I need this to formulate my program gofchi2 mor generally.

*

* For searches and help try:

* http://www.stata.com/support/faqs/res/findit.html

* http://www.stata.com/support/statalist/faq

* http://www.ats.ucla.edu/stat/stata/

Nick Thank you very much for prompt answer. Wolfgang

Nick's strategy to extract the varname from the e(wexp)-container works very fine, but I need for further calculations the content of the variable - its values in a vector of 8 subpopulations. How can I get back form the string expression extracted out of e(wexp), the name of the variable used for weighting, to its vector?

Thank you in advance

Wolfgang

--

Dr. Wolfgang Langer

Martin-Luther-Universität Halle-Wittenberg

Institut für Soziologie

Emil-Abderhalden-Str. 7

06099 Halle (Saale)

Martin-Luther-University Halle-Wittenberg

Institute of Sociology

Emil-Abderhalden-Str. 7

D06099 Halle (Saale)

E-mail: wolfgang.langer@soziologie.uni-halle.de

Homepage: www.soziologie.uni-halle.de/langer/

Institute:

Tel.: ++49/ (0)345 5524248

Fax.: ++49/ (0)345 5527149

Privat:

Tel.: ++49/ (0)345 2021819

"It's better to have an approximate solution to the right problem

than to have an exact solution to the wrong one." (John Tukey 1988)

or for insiders

"Statistics can't work magic." (Howard Winer 1989)

or for optimists

"Murphy's Laws for Researchers:

(1) If you think of something new, it's been done. (2) If you think something is important, no one else will. (3) If you throw it away, someone else will publish it, obtain a grant, write a book, and get on the Oprah Winfrey show." (Joop Hox 2000)

*

* For searches and help try:

* http://www.stata.com/support/faqs/res/findit.html

* http://www.stata.com/support/statalist/faq

* http://www.ats.ucla.edu/stat/stata/

**References**:**Re: st: Extracting the varname of e(wexp) after regress ?***From:*n j cox <n.j.cox@durham.ac.uk>

**Re: st: Extracting the varname of e(wexp) after regress ?***From:*Wolfgang Langer <langer@soziologie.uni-halle.de>

- Prev by Date:
**RE: st: Help with Reading Arguments for Do-File** - Next by Date:
**RE: st: Extracting the varname of e(wexp) after regress ? Another question more** - Previous by thread:
**Re: st: Extracting the varname of e(wexp) after regress ?** - Next by thread:
**st: file write handle?** - Index(es):

© Copyright 1996–2017 StataCorp LLC | Terms of use | Privacy | Contact us | What's new | Site index |