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.
Asked
Active
Viewed 66 times
0
-
See also [What is the difference between linear regression and logistic regression?](http://stats.stackexchange.com/q/29325/17230). – Scortchi - Reinstate Monica Mar 04 '16 at 12:43
-
No but that answer is not satisfying me – Milan Amrut Joshi Mar 04 '16 at 12:45
-
@ Scortchi kindly see the question again – Milan Amrut Joshi Mar 04 '16 at 12:48
-
1See [Problem understanding the logistic regression link function](http://stats.stackexchange.com/q/80611/17230); also [Difference between logit and probit models](http://stats.stackexchange.com/q/20523/17230), & [Is the logit function always the best for regression modeling of binary data?](http://stats.stackexchange.com/q/48072/17230). 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 $\operatorname{E} (Y\mid x) = \Pr(Y=1 \mid x)$? – Scortchi - Reinstate Monica Mar 04 '16 at 13:22