Can you please, do not write my question on the internet server,
Thanks
De : Gubelmann Carine
Envoyé : mardi, 16. mars 2010 13:36
À : 'r-h...@stat.math.ethz.ch'
Objet : [R] G-test : log-likelihood ratio test
Dear Peter,
I am running your g.test() with the william's correction but I have a question
about the input numbers. These are my data:
"Our data are consistent with those obtained using microarray comparative
genome hybridization in that we found significantly fewer variants per Mb on
the X compared to the autosomal chromosomes (152 versus 336 respectively, G =
93.4, P < 2e-16, df = 1), and in exons (including UTRs) compared to non-exonic
regions (69 versus 392 respectively, G = 14,226, P < 2e-16, df = 1)."
How do I have to write the input? Should I have to include any total number (in
this case Mb) minus the variants?
g.test(rbind(c(152,(999848)),c(336,(999664))), correct = "williams") or
g.test(rbind(c(152,336)), correct = "williams") ?
Thanks a lot for your help,
Carine
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
