As stated in this course pdf, Gaussian Discriminant Analysis (GDA) can also be expressed in the form of $\frac{1}{1+\exp(-\theta^Tx)}$, where $\theta$ is some appropriate function of $φ$, $\Sigma$, $u_0$, and $u_1$. So why can't logistic regression predict model like them?
$\begingroup$
$\endgroup$
2
-
$\begingroup$ What do you mean by "why can't logistic regression predict model like them"? What does the "them" imply? $\endgroup$– Lerner ZhangCommented Jun 24, 2018 at 10:39
-
$\begingroup$ Take a look at this question stats.stackexchange.com/questions/251069/… $\endgroup$– Daniel ChepenkoCommented Jun 25, 2018 at 2:22
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
GDA makes more specific assumptions about the data set then Logistic Regression and if those assumptions are true then it works better then LR. But on the other hand LR makes more generic assumptions and can be more useful in lot of other places where the probability distribution of the feature set is not Gaussian.
