In OLS regression we estimate the parameters of the Linear regression with multiple variables using the "Normal Equation":
Would it be possible to do something similar when estimating the parameter for logistic regression?
Asked
Active
Viewed 1,256 times
3
-
I think you're making an analogy to the normal equations in OLS, right? To make it crystal clear what you mean, I suggest you start your question "In OLS regression we estimate parameters by solving the "normal equations" (then quote the normal equations). This gives a little bit more context for what you're asking. – Silverfish Nov 12 '15 at 11:16
-
2You didn't do as Silverfish suggested, since you didn't then quote the normal equations. – Glen_b Nov 12 '15 at 14:00
-
see chapter 3.2.1 of this link http://data.princeton.edu/wws509/notes/c3.pdf – Jan 23 '16 at 16:54