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

From |
"Nick Cox" <[email protected]> |

To |
<[email protected]> |

Subject |
RE: st: Programming question |

Date |
Thu, 30 Oct 2003 14:30:24 -0000 |

Winston's do file starts (omitting stuff which looks irrelevant or unproblematic) set obs 200 genlev 5 10 centre genlev 4 10 strata set seed 745363 gen rand1=uniform() gen rand2=uniform() genlev 2 1 treat sort centre strata block * making rand2 constant within each block quie by centre strata block: replace rand2=rand2[_n-1] if _n~=1 * sorting blocks in random order and * sorting treatments randomly within blocks sort centre strata rand2 rand1 genlev 20 1 pid genlev 4 10 dose genlev 4 10 age genlev 2 20 sex Even on Winston's guess I am unclear how the numbers fit together for a dataset of 200. A secondary question is that if -genlev- involves some generation of random numbers then the -set seed- statement belongs earlier. Two of my earlier suggestions remain in the air. Winston Banya > MY GUESS is that genlev 5 10 centre and genlev 4 10 strata would mean > generate (randomly) 5 centres in blocks and four strata in blocks with each > block containing 10 people. The four strata are Males 50-69 years, Males > 70+ Females 50-69 and females 70+. Within each stratum there are equal > number of treatments (i.e 1 & 2) I do not know who wrote the program. I only > inherited it when I got my present job as it was on the computer that was > assigned to me. > > What I need is a randomly generated list fitting the brief description above > but I do not know how to do produce it through writing a program like the > one I inherited thus the reason for posting the question on statlist. Nick Cox > Your problem is that your .do file uses a -genlev- command > which is not part of official Stata, was not published in > the STB or the SJ, and is not available from SSC or other > prominent Stata websites. Or at least > I cannot find -genlev- using either -search- or its superset > -findit-. > > Your options include the following: > > * Perhaps -genlev- is lurking somewhere on your machine > or your system, just not visible to the Stata you are > running. Get your OS to look for it. If you find it, > copy it to somewhere along your -adopath-. You are > looking for a file called genlev.ado. > > * It sounds as if -genlev- was written by some present > or past colleague. Ask the local Stata expert where it came > from. If you are the local expert, ask a bigger less local > Stata expert. Or here we are on Statalist, so does anyone > have a -genlev- on their system, i.e. does > > . which genlev > > find a file? > > * Tell us exactly what -genlev- does, and it's probably > easy then to write a substitute. Personally I can't > reconstruct what -genlev- does from the syntax used > to call it. * * 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/

- Prev by Date:
**Re: st: RE: Log Likelihood for Linear Regression Models** - Next by Date:
**st: legend of -graph pie-** - Previous by thread:
**RE: st: Programming question** - Next by thread:
**st: programming question** - Index(es):

© Copyright 1996–2024 StataCorp LLC | Terms of use | Privacy | Contact us | What's new | Site index |