Given the formula $OR=frac{P_1(1-P_2)}{P_2(1-P_1)}$ for calculating the odds ratio ($OR$). How can I calculate the standard error of $OR$ based on the standard errors of $P_1$ and $P_2$?
Contents
hide
Best Answer
If you have the cell counts you can get the standard error for the $ln(OR)$ (see for exampel here) :
$sqrt{frac{1}{n_{11}}+frac{1}{n_{10}}+frac{1}{n_{01}}+frac{1}{n_{00}}}$
Using the delta method, you can approximate the standard error of the $OR$ as:
$OR times sqrt{frac{1}{n_{11}}+frac{1}{n_{10}}+frac{1}{n_{01}}+frac{1}{n_{00}}}$
Similar Posts:
- Solved – How to calculate SE of an odds ratio
- Solved – Does confidence interval for odds ratio assume log-normal distribution
- Solved – Calculating standard error of log response ratios from standard errors of the two means
- Solved – How to calculate odds ratio and p-values for Fisher’s test
- Solved – How to calculate odds ratio and p-values for Fisher’s test