Dear Jon,
I would suggest you to use logarithmic transformation of the proportions of male and female. Do it in following way:
P1=Proportion of male
P2=Proportion of female
LnP1=log(P1)
LnP2=log(P2)
Var(LnP1-LnP2)=Var(LnP1) + Var(LnP2) - 2*Cov(LnP1, LnP2)
SE(LnP1-LnP2)=square root of Var(LnP1-LnP2)
C.I. for P1/P2 will be:
exp[(LnP1-LnP2)+/- SE(LnP1-LnP2)]
I hope it helps you.
Best regards
MADAN GOPAL KUNDU
"Jon Heron (ALSPAC)" <[log in to unmask]> wrote:
Dear all,
I want to calculate the CI for a sex-ratio within my dataset
(to examine evidence against it being 1:1).
I have calculated an estimate of the proportion of boys with a CI
for this and it would be trivial matter to calculate a sex-ratio from this
estimate. It seems reasonable I should be able to transform the CI in the
same way - is this OK?
I'm torn between "yes, of course it is" and "hmm, that seems a bit
too easy"
cheers
Jon
--------------------------------------------------
Dr Jon Heron
Statistics Team Leader
ALSPAC, Dept of Social Medicine
24 Tyndall Avenue
Bristol BS8 1TQ
Tel: 0117 3311616
Fax: 0117 3311704
--------------
Madan Gopal Kundu
Biostatistician and SAS Programmer
Ranbaxy R&D
Gurgaon, Haryana
India
Web: http://www.freewebs.com/madanstata
mobile: 91-9868788406
e-mail: [log in to unmask]
Click to join Statisticians_group
---------------------------------
Find out what India is talking about on - Yahoo! Answers India
Send FREE SMS to your friend's mobile from Yahoo! Messenger Version 8. Get it NOW
|