Hi! I just wanted to point out that in the process of creating the confidence interval for the odds ratio it is incorrectly applied (maybe un purpose? and just missing a note(?) IDK) as the sum of square roots is rarely the square root of a sum. You get the correct result but the method is not correct. You can see here https://math.mit.edu/~rmd/650/deltamethod.pdf for a correct derivation of the OR. You can also do it with the gradient and variance covariance matrix directly (as in the book Asymptotic Statistics). Let me know if you need help, as a statistician I am happy to fill any blank missing ;)
Hi! I just wanted to point out that in the process of creating the confidence interval for the odds ratio it is incorrectly applied (maybe un purpose? and just missing a note(?) IDK) as the sum of square roots is rarely the square root of a sum. You get the correct result but the method is not correct. You can see here https://math.mit.edu/~rmd/650/deltamethod.pdf for a correct derivation of the OR. You can also do it with the gradient and variance covariance matrix directly (as in the book Asymptotic Statistics). Let me know if you need help, as a statistician I am happy to fill any blank missing ;)