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

From |
"Nick Cox" <n.j.cox@durham.ac.uk> |

To |
<statalist@hsphsun2.harvard.edu> |

Subject |
RE: st: RE: How to find Max value. |

Date |
Sat, 19 Jun 2004 20:22:42 +0100 |

Thank you for the further detail. However, if I understand correctly, you are asking a very strange question here. Since -mf- (presumably -rf- is the same thing or similar) is measured to 1 or 2 decimal places, the maximum is quite likely to be unique for each subset (time and temperature), and the percent equal to the maximum will just be 100 / number in subset. Fortuitously, you may get tied values, but it is difficult to believe that their frequency would reflect properties of statistical or scientific interest. In addition, if you attempt to count how many values are equal to the maximum, you are likely to run into the precision problems detailed at [U] 16.10. Providing Stata code would just be giving you the means of doing something bizarre, I believe. Nick n.j.cox@durham.ac.uk Syed O Masood > I have put a piece of data below. > I need to find what percent of people have maximum > value in the variable rf at time 1 and temp 26, what > percent of people will have maximum value of mf at > time 2 and temp 26, what percent of people have max > value of mf at time 3 and temp 26 and similarly for > tmep 35oC. > > Thanks in advance for your help & support > > . list temp time mf > > +---------------------+ > | temp time mf | > |---------------------| > 1. | 26 1 5.5 | > 2. | 26 2 8.64 | > 3. | 26 3 12.03 | > 4. | 35 1 8.64 | > 5. | 35 2 5.59 | > |---------------------| > 6. | 35 3 9.02 | > 7. | 26 1 3.88 | > 8. | 26 2 8.63 | > 9. | 26 3 10.35 | > 10. | 35 1 8.34 | > |---------------------| > 11. | 35 2 10.22 | > 12. | 35 3 11.25 | > 13. | 26 1 3.29 | > 14. | 26 2 6.12 | > 15. | 26 3 7.37 | > |---------------------| > > --- Nick Cox <n.j.cox@durham.ac.uk> wrote: > > This isn't very clear to me. What percent > > of anything is equal to the maximum value > > is very sensitive to the resolution of measurement, > > and as such doesn't seem a very obvious > > thing to calculate. > > > > You may get a better answer if you give > > an example table showing some data. > > > > Nick > > n.j.cox@durham.ac.uk * * 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: How to find Max value.** - Next by Date:
**RE: st: RE: How to find Max value.** - Previous by thread:
**RE: st: RE: How to find Max value.** - Next by thread:
**RE: st: RE: How to find Max value.** - Index(es):

© Copyright 1996–2015 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |