I am new to this area of regression analysis
why do we select ${e^{\beta x } }/{[1+e^{ \beta x}]}$ as $p\{Y=1|X\}$
or how to get $E(y |x )$ in the case where $y$ is dichotomous.
$\begingroup$
$\endgroup$
4
-
$\begingroup$ See also What is the difference between linear regression and logistic regression?. $\endgroup$– Scortchi ♦Commented Mar 4, 2016 at 12:43
-
$\begingroup$ No but that answer is not satisfying me $\endgroup$– Milan Amrut JoshiCommented Mar 4, 2016 at 12:45
-
$\begingroup$ @ Scortchi kindly see the question again $\endgroup$– Milan Amrut JoshiCommented Mar 4, 2016 at 12:48
-
1$\begingroup$ See Problem understanding the logistic regression link function; also Difference between logit and probit models, & Is the logit function always the best for regression modeling of binary data?. Not sure what you mean by "how to get $\operatorname{E} (Y\mid x)$" - performing logistic regression is a way to estimate it. Or are you asking why $\operatorname{E} (Y\mid x) = \Pr(Y=1 \mid x)$? $\endgroup$– Scortchi ♦Commented Mar 4, 2016 at 13:22
Add a comment
|