1
$\begingroup$

In simple logistic regression i noticed that many researcher neglect the constant part of the logistic regression equation.

Log(odds) = a + b Gender (say that 0 = Female and 1 = Male) So the log of the odds of cancer when the case is male is b ... and odds of course is exp(b)

but shouldn't the log of the odds equals b + a? because log(odds) = a + b x 1 = a + b

$\endgroup$
1

1 Answer 1

0
$\begingroup$

My guess is that you've misinterpreted what someone wrote and they meant that males have $b$ higher log-odds than females.

$$ \log(odds) = a+bx\\ \log(odds) = a+b\text{ (male)}\\ \log(odds) = a+0\text{ (female)}\\ $$

Then subtract to see that $\log(odds)$ for males is $b$ higher than $\log(odds)$ for females.

$\endgroup$
1
  • $\begingroup$ Yeah I understand it like what you said. But I in many articles it's just interpreted as: odds of cancer for males is exp(b), without taking care of a $\endgroup$
    – Hussain
    Feb 1, 2022 at 15:53

Your Answer

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

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