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

From |
Richard Williams <Richard.A.Williams.5@nd.edu> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: - uniform() - random variable not uniformly distributed overinterval |

Date |
Wed, 09 Mar 2005 20:53:24 -0500 |

At 05:19 PM 3/9/2005 -0800, Deborah Garvey wrote:

Actually, it is very unlikely that a random variable will produce exactly a 50/50 split. If you repeated this process several times, you would find that sometimes the first group was bigger, sometimes the 2nd group was bigger. By chance alone, sometimes the mean and median will be greater than .5, sometimes they will be less.Dear listers: I am generating a random number with the goal of (randomly) assigning children who report multiple races to a single race category for purposes of analysis. I'd like to evenly distribute children across single race categories. However, when I use uniform() to generate a random number, I don't get an even split: ... I am puzzled. I would expect median = mean = 0.5 for a uniform number defined over [0,1). This must be a simple, basic question, but neither changing the - set seed -, changing the interval to [0,1], or changing the observations selected seems to solve this problem. A search of on-line help, the list archives and the Stata FAQs didn't give me much insight either.

There are various solutions. If you don't have it already, I would suggest installing -egenmore- (give the command -findit egenmore- to find and install.) Then, give the command

egen r = rndsub() if k_race == 801, ng(2)

If I've done this right, I believe you'll get 1155 in your first group and 1154 in your 2nd. You should probably precede this with a -set seed- command so you can replicate your results, e.g.

set seed 12345

egen r = rndsub() if k_race == 801, ng(2)

replace k_race2 = 100 if r = 1

replace k_race2 = 200 if r = 2

-------------------------------------------

Richard Williams, Notre Dame Dept of Sociology

OFFICE: (574)631-6668, (574)631-6463

FAX: (574)288-4373

HOME: (574)289-5227

EMAIL: Richard.A.Williams.5@ND.Edu

WWW (personal): http://www.nd.edu/~rwilliam

WWW (department): http://www.nd.edu/~soc

*

* 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/

**Follow-Ups**:**Re: st: - uniform() - random variable not uniformly distributed overinterval***From:*Chris Ruebeck <ruebeckc@lafayette.edu>

**References**:**st: - uniform() - random variable not uniformly distributed overinterval***From:*"Deborah Garvey" <DGarvey@scu.edu>

- Prev by Date:
**Re: st: - uniform() - random variable not uniformlydistributed overinterval** - Next by Date:
**Re: st: - uniform() - random variable not uniformly distributed overinterval** - Previous by thread:
**st: - uniform() - random variable not uniformly distributed overinterval** - Next by thread:
**Re: st: - uniform() - random variable not uniformly distributed overinterval** - Index(es):

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