Hi All, This is not directly related to R but I post the questions here since there are a lot of experts on statistics. I want to calculate power of logistic regression using likelihood ratio test in unmatched case control design. The paper I have read is "power calculations for likelihood ratio tests in generalized linear models" by Steven G. Self in 1992. They showed a formula for calculating power for prospective cohort design. Can I plug in the various parameters from case control study? That is, can I pretend that the case control data I have is a prospective cohort study? I guess I can do this since in principal logistic regression is valid for both case control and cohort design in estimating odds ratio. Thanks a lot!