1
$\begingroup$

I am reading a paper on how the authors calculate the variance/standard deviation of what appears to be log(odds).

The paper is a medical paper (Discontinuation of Oral Antivirals in Chronic Hepatitis B: A Systematic Review).

In the paper, it seems that they are doing something equivalent of log(odds) by using $\log(p/(1-p))$.

However, I don't understand how they get the variance to be $1/[Np(1-p)]$.

I am confused by other sources stating that the standard deviation of log odds is supposed to be $\sqrt{1/a+1/b+1/c+1/d}$. For instance, How do I calculate the standard deviation of the log-odds?.

Are the two approaches the same or different? Which is correct?

cited text as pdf

$\endgroup$

1 Answer 1

3
$\begingroup$

$$\hat Var(log(\frac p{1-p})) \approx \frac 1 {Np(1-p)}$$

$$\hat Var\left[log\left(\frac {p_1/(1-p_1)}{p_2/(1-p_2)}\right)\right] \approx \frac 1 a + \frac 1 b+ \frac 1 c +\frac 1 d$$ where $p = X/N$ is estimate of parameter $\pi$ in $X\sim Bin(\pi,n)$

First one is for log ODDS, and second one is for log ODDS RATIO. In the linked page, OP misused ODDS.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.