Sample size tables for logistic regression

摘要: Sample size tables are presented for epidemiologic studies which extend the use of Whittemore's formula. The easy to both simple and multiple logistic regressions. Monte Carlo simulations performed show three important results. Firstly, sample suitable with either high or low event proportions. Secondly, although can be inaccurate risk factors having double exponential distributions, they reasonably adequate normal distributions distributions. Finally, power a study varies number events individuals at risk.