|
Post by shirleydai on Jul 28, 2017 3:02:25 GMT
Hello, I'm pretty new here. When I read GCTA user manual for COJO analysis, I found a note said "For a case‐control study, the effect size should be log(odds ratio) with its corresponding standard error". Previously I got GWAS association data from PLINK with OR and its standard error (SE). I wonder how can I convert this SE here, when after converting OR to log(OR)? Thanks Shirley
|
|
|
Post by Jian Yang on Jul 30, 2017 10:27:02 GMT
Z = log(OR) / SE. So, SE = log(OR) / Z where Z can be computed from p-value.
|
|